In Defense Of Drafting For Need
by onlxn on Jun 23, 2009 11:53 PM PDT
50 comments
This time of year, both of GSOM and on most other NBA blogs, you'll see the acronym "BPA" an awful lot. The credo of "best player available" is a fairly sacred one in fan circles, and understandably so. If your team is bad -- and the teams of most fans who obsess over the draft surely are -- you want your team to aggregate talent above all else. Don't worry about making pieces fit if the resulting puzzle is ugly. Don't pass up Michael Jordan just because you have Clyde Drexler; don't draft Patrick O'Bryant just because you want a center and he's the next one on the list. When drafting, it is said, your team should follow one and only one commandment: DRAFT THE BEST PLAYER AVAILABLE.
That's wrong.
It's not wrong in the simplest sense -- you *should* draft the best player available whenever possible. But the point is, *there is usually no obvious best player available*. This year, anyone with a pulse will tell you that the best player is Blake Griffin, and the Clippers will draft him accordingly. But that's the exception, not the rule. Memphis currently sits at two... the consensus pick there is probably Ricky Rubio, but if he's a majority choice, he's only barely one. The third pick is probably Hasheem Thabeet by a weak plurality, and after that it's pure chaos. In well over 90% of NBA draft slots, this year and in most years, nobody agrees on who the best player is.
That's not to say that a particular person doesn't have a strong opinion about the best available players. Many GSOM posters have ranked the top prospects in some order or another... NBA GMs obviously do the same. But it's unrealistically reductive to assume that every GM's top-ranked guy is and should be the right choice. GMs are surrounded by assistants, coaches, scouts, statisticians, owners, friends, mentors, influential and knowledgeable players... the best GMs listen to these people. Do the GMs ultimately reject the input of others and take the guy they had in their gut all along? Sure... they often do. But they sometimes don't, and they're sometimes right to stop listening to their gut. Determining who the best available player is is often next to impossible. And it's sort of obvious why.
Think about it. You're evaluating hundreds of players, some as young as 19, some (in the case of certain overseas players) as old as 25. Players from dozens of different countries, players of all different positions and shapes and sizes. You're judging them on play -- shooting, driving, rebounding, passing, defense -- but if you're a good GM, you're looking at a lot more than that. You're looking at their psychological makeup. You're looking at their health. You're looking at their chances to be a superstar... you're looking at their chances to be a bust. You're looking at their ability to handle the media, and yes, you're probably looking at their marketing potential, or lack thereof. The NBA is an infinitely complicated business. It's tempting to say that the right drafting strategy is determining the best available player and picking that person, but the situation is often not nearly so simple. Quite often, you're left with a bunch of guys who, when you balance all the various dimensions, seem about equally promising. If that's the case, how do you break the tie?
It's really sort of obvious: you draft the guy you *need*.
There are several relevant actors in the Kings' front office, and they may not -- by most accounts, do not -- agree on the best player who'll be available at four. But they can agree on one thing: *they need a point guard*. If all else is equal -- if you can draft a B-minus power forward or a B-minus point guard -- why *wouldn't* you draft the guy who'd actually address the issues on your basketball team?
Many Warriors fans, including many of the smartest Warriors fans, are allergic to this idea. When you have a crappy team like ours, the feeling is that you need to maximize the potential of every single draft pick. Don't let a stud pass you buy... don't let a dud seduce you because of fit. And at the extremes, this philosophy is absolutely right. If the available picks include Danny Granger and a bunch of guys who have significantly less potential than Danny Granger, forget about positions and take Danny Granger. If you really want a center, but the best one available is significantly worse than many available players at other positions, you shouldn't draft a center. I don't think anyone here would disagree.
But usually, it's not nearly that clear. Usually, you're looking at a bunch of guys with comparably similar potential. In that scenario, not only *can* you think of your team's needs, you *must*. Which is why the ideal drafting philosophy is *not* as simple as "BPA"... it's something more like this:
1) If one player is *clearly* the best available player, you must take that player.
2) If no such player exists, you must take the top-tier player who best fits your team's needs.
I don't always (or even often) agree with Chad Ford, but his concept of draft "tiers" is absolutely right. He has revealed his rankings of this year's top crop. They are by no means the gospel, and I actively disagree with a couple of his rankings, but let's regard them as wise for the sake of argument.
Tier 1
Tier 2
James HardenRicky Rubio
Hasheem Thabeet
Tier 3
Stephen CurryDeMar DeRozan
Tyreke Evans
Jonny Flynn
Jordan Hill
Jrue Holiday
What does this mean? It means that if Blake Griffin is available, you *have* to take him, no matter your positional situation, because he's the only Tier 1 guy there is. (Even the Clippers understand that.) But if you have the #2 pick -- whether you're Memphis, Minnesota or whoever -- it's not quite so simple. There are three available players of fairly similar value, similar enough so that trying to determine the "best player available" is essentially impossible. In that case, take the Tier 2 guy who fits you best. For the Grizzles, it's probably Thabeet... for the Wolves, it's probably Rubio. Either pick would be completely defensible, even if you thought a guy who fit you worse might have *mildly* higher potential.
The Warriors? Well, as always, we're a weird case. It's not actually that clear what our "need" position even is. *But it is totally appropriate that we would take that into consideration.* If there is no available guy who jumps out, you *have* to take the shape of your roster and franchise into account. There is no reward at year's end for the team that accrued the most players with agreed-upon "potential" scores... there is only reward for the teams that won the most basketball games. Fit shouldn't trump potential, especially on bad teams... but fit is far easier to measure than potential. When the potential of two players seems for all the world to be similar, you should pick the one you think you *need* more. That's just logic.
I don't post this to suggest we should be leaning hard towards one player or away from another... I don't follow the college game at all closely enough, or understand the difference between levels well enough, to have an informed opinion about any of these guys. But I do think our collective conversation could use a little kick in the pants. "BPA" is not a very illuminating credo by itself. Need is of secondary but still-critical importance, and any argument that omits that misses the boat.
This FanPost is a submission from a member of the mighty Golden State of Mind community. While we're all here to throw up that W, these words do not necessarily reflect the views of the GSoM Crew. Still, chances are the preceding post is Unstoppable Baby!
5 recs |
50 comments
Email
Print
Comments
i would take need over best player available
by gsw.raiders on Jun 24, 2009 12:01 AM PDT reply actions 0 recs
Thing is, when you suck, as with most teams picking at the top of the draft, your “need” is more talent.
Can you point to even one situation in which picking the BPA proved, short term or long term, to be an unwise move? Cos I can point to a ton in which picking for STPPN — “short-term perceived positional need” — proved disastrous (though I suppose one can never prove a pick was based on STPPN).
I dunno, onlxn: I basically agree with a lot of what you’ve written; at the same time I think you’ve helped to muddy the waters of a pretty simple argument that for some reason still confuses a lot of fans.
Most would agree that shooting guard is currently the least of our “needs.” If (young) Jordan were in this draft and we had the #1 pick, would you pick him? Does this issue have to be any more complicated than that?
Thing 1
by Sleepy Freud on Jun 24, 2009 2:14 AM PDT up reply actions 0 recs
Patrick Obryant
Todd Fuller
Ike Diogu
Adonal Foyle
We don’t need any more of those guys.
LeBron James? I'm the only Ty Crane.
by misterjennings on Jun 24, 2009 7:27 AM PDT up reply actions 0 recs
Rondo
Kobe
Granger
TMac (when he was young and healthy)
We need more of those guys.
Thing 1
by Sleepy Freud on Jun 24, 2009 7:41 AM PDT up reply actions 0 recs
Can you point to even one situation in which picking the BPA proved, short term or long term, to be an unwise move?
Um, Marvin Williams over Deron Williams and Chris Paul? The Hawks were heavy at the three and badly needed a point guard, but were so convinced of Williams’s special gifts that they took him anyway, need be damned. In so doing, they muddied their roster with a glut of swings (a problem that helped to lose Josh Childress) and missed out on an All-Star and one of the best point guards in league history.
How about Darko Milicic over Carmelo Anthony? The Pistons could’ve chosen an athletic scorer who would’ve helped them right away… instead, they went for the big project, thinking he had the bigger upside. We’re still waiting on that upside… meanwhile, Melo’s a gold medalist who just led his team to the Final Four.
Now, you can argue that thesee teams incorrectly identified who the best available player was. But you don’t get to make drafts retroactively. Of course picking the actual best player available is the right move, but it’s pretty rare that teams can know who that guy is. Both the Marvjn Williams pick and the Darko pick had a number of vocal supporters, and those supporters’ mantra was BPA.
Most would agree that shooting guard is currently the least of our "needs." If (young) Jordan were in this draft and we had the #1 pick, would you pick him? Does this issue have to be any more complicated than that?
Of course it does. 99.99% of the picks in NBA history have not been Michael Jordan… you almost never draft in a slot where there’s an obvious standout. If your draft strategy is as simple as “draft the guy with the best chance to be Michael Jordan”, you’re going to end up with a lot of Gerald Greens and Tim Thomases and never get anywhere.
Think about it this way: let’s say we even do know what these players’ careers will look like ahead of time. Let’s say we’re trying to decide between Stephen Curry and Jordan Hill. A magic ball tells us that Stephen Curry will have, say, Mark Price’s career, and Jordan Hill will have Anthony Mason’s career… both guys will be pretty good players for a dozen years. You’re telling me that you shouldn’t give any thought to whether you’d rather have Mark Price or Anthony Mason? You just have to take whichever player did infinitesimally better in their NBA career, fit be damned?
No. You take anybody who’s head-and-shoulders above the rest… beyond that, you can — should — must — think about fit.
by onlxn on Jun 24, 2009 9:07 AM PDT up reply actions 0 recs
I mean, you kind of made the point for us, that they misidenfitied who the best player was and got too caught up in “potential” without thinking about the likelihood of reaching that potential. Hawks should have picked Chris Paul, and what if they later picked Rondo. Sure they already have Paul (hypothetically), but having 2 talented players isn’t a bad thing. Jae made the point that players can get bumped up higher than they should be on a draft board based on potential, without enough consideration for the likelihood that they reach it, which is true (and a BPA mentality might promote this). It’s also true that sometimes players can get bumped up higher than they should be on a draft board based on position, without enough consideration for their talent (and a need mentality will promote this).
Ultimately this is why I like the tier system, it allows you to go for the best overall talent, while also drafting for need if there are comparable talents. It’s up to the scouts to determine each players talent level, potential, and likelihood of reaching their potential, and often times they fail at this aspect (predicting the future is difficult, after all).
For a bad team (like a team picking around…#7), our primary goal should be acquiring as much talent as possible now, and figuring out what to do with it in the future.
by Missing Barry on Jun 24, 2009 9:57 AM PDT up reply actions 0 recs
Um, Marvin Williams over Deron Williams and Chris Paul?
That was a simple case of being wrong about who the BPA actually was. If Marvin was actually better than Deron Williams and Chris Paul that would have been a great pick. When you make a bad judgment about a player you can’t really blame your draft strategy, that’s just bad judgment (or scouting).
How about Darko Milicic over Carmelo Anthony?
When Dumars is asked about that pick he always says that a big part of his decision making was the fact that he already had Prince at SF. Which is an absolutely great example of picking for need instead of picking the BPA.
Nobody has a perfect draft record, but you can’t throw a solid strategy under the buss because teams have a hard time actually identifying the BPA. That should always be the goal.
Thing 2
by olympicmike on Jun 24, 2009 10:05 AM PDT up reply actions 0 recs
you can’t throw a solid strategy under the buss
Isn’t that what the Lakers did with Kobe?
Thing 1
by Sleepy Freud on Jun 24, 2009 10:48 AM PDT up reply actions 0 recs
oops...
I do think that Phil Jackson has spent plenty of time under the Buss…
Thing 2
by olympicmike on Jun 24, 2009 5:46 PM PDT up reply actions 0 recs
You have been DFiBrillated.
by Dubs fan in Boston on Jun 30, 2009 11:37 AM PDT up reply actions 0 recs
Um, Marvin Williams over Deron Williams and Chris Paul?
As you say, in bold, “you don’t get to make picks retroactively.” The error was in misjudging who the best player was, not in the approach of BPA. Had the Hawks had a solid starting PG in place, and no one decent at the wings, picking Paul still would have been the correct move.
How about Darko Milicic over Carmelo Anthony?
I would argue that this was an example of picking for need (since the Pistons were pretty well set at the wings) rather than BPA. You’re conflating BPA with the dreaded “DOP” — drafting on potential — and thereby constructing a bit of a strawman. You’re right to point out that the BPA approach is not as simple as I make it out to be, in that it combines an assessment of “upside” with one of “realistic chance of attaining that upside.” Not being a stats guy, I’m not sure, mathematically, how this combination is arrived at, but I suspect it’s something like the product of these two values.
Think about it this way: let’s say we even do know what these players’ careers will look like ahead of time. Let’s say we’re trying to decide between Stephen Curry and Jordan Hill. A magic ball tells us that Stephen Curry will have, say, Mark Price’s career, and Jordan Hill will have Anthony Mason’s career… both guys will be pretty good players for a dozen years. You’re telling me that you shouldn’t give any thought to whether you’d rather have Mark Price or Anthony Mason? You just have to take whichever player did infinitesimally better in their NBA career, fit be damned?
All this seems needlessly complex and again, a bit strawman-ish. If, using the BPA system, you project Curry to be a better player in terms of his contribution to wins (adjusting of course for the fact that he plays a less rare position), you pick Curry. Even if you feel you’re “thin” at PF. Are you really suggesting otherwise?
Thing 1
by Sleepy Freud on Jun 24, 2009 10:14 AM PDT up reply actions 0 recs
[Edit: damn you, Thing 2!]
Thing 1
by Sleepy Freud on Jun 24, 2009 10:15 AM PDT up reply actions 0 recs
Haha
The similarity in format is almost as creepy as the similarity in content.
Thing 2
by olympicmike on Jun 24, 2009 10:39 AM PDT up reply actions 0 recs
As you say, in bold, "you don’t get to make picks retroactively." The error was in misjudging who the best player was, not in the approach of BPA.
It’s an error a lot of people made, though. Many, many observers praised the Hawks for that pick — most analysts had pegged Williams as a better player than Paul, not by a lot, but a little. Following a pure “BPA” strategy with the Hawks’ pick, at the time, would have probably led you — and did lead them — to Marvin Williams.
Following a tiering strategy would have led you to Deron or CP3. The evaluations of those two and Marvin Williams were close enough at the time that there was no real consensus as to who’d be the best player of the three. In that scenario, where it’s more or less a tie, need can and should come into play… the Hawks would’ve picked one of the points.
Had the Hawks had a solid starting PG in place, and no one decent at the wings, picking Paul still would have been the correct move.
No doubt… if you can grab one of the best point guards ever, it’s worth shuffling things around. And in a tiering scenario, a PG-heavy Hawks team would’ve drafted Marvin Williams and missed out on CP3 anyway. There are risks and costs to any drafting strategy. But saying “the Hawks should’ve known Paul was the BPA and taken him” is not an endorsement of one drafting philosophy over another… that’s an argument for seeing the future. Nobody can project player rankings perfectly, not even Jerry West. What you can project quite well is how your team fits together.
I would argue that this was an example of picking for need (since the Pistons were pretty well set at the wings) rather than BPA.
It’s a fair point… it’s true that Dumars did try to justify it that way. My perspective is that if you’re a title contender, the “need” pick would be a guy who could contribute right away, not a project. “Need” can cut in many ways.
You’re conflating BPA with the dreaded "DOP" — drafting on potential — and thereby constructing a bit of a strawman.
Mm, I’m really not seeing the strawman here.
Who are the two best players we’ve passed over in the past fifteen years? KG and Kobe. Both were high schoolers, taken at a time when people were very scared of taking chances on those guys… no high schooler had been drafted in fifteen years. KG, also, was skinny as a rail. Those guys were Drafted On Potential, just like Darko. There were risks — big ones, in the case of Garnett. You can not separate the “BPA” misses from the “BPA” hits. You can’t retroactively chalk up every bad draft pick to a philosophical mistake. Most picks ever made by NBA teams have been “BPA”, and many, many, many of them have been wrong. No GM, however smart, has ever been able to consistently pick the best player available. Don’t you think there’s an argument for letting some other thoughts into the conversation, when there’s a tie to break?
You’re right to point out that the BPA approach is not as simple as I make it out to be, in that it combines an assessment of "upside" with one of "realistic chance of attaining that upside." Not being a stats guy, I’m not sure, mathematically, how this combination is arrived at, but I suspect it’s something like the product of these two values.
I’d think a formula would have to be more complicated than that — you’d want to include value in downside scenarios, health, age, likely development speed, etc. Really, I’d doubt that anyone has a codified formula they use to give possible draftees a single numerical value. But besides the point, I suppose.
If, using the BPA system, you project Curry to be a better player in terms of his contribution to wins (adjusting of course for the fact that he plays a less rare position), you pick Curry. Even if you feel you’re "thin" at PF. Are you really suggesting otherwise?
If the two players are extremely close in terms of value? Yes.
Price and Mason were both good-but-not-elite players who played for about a dozen years… their total contributions weren’t far off in value. There’s no need to walk through their actual careers, but let’s say, for the sake of argument, that Anthony Mason was worth twelve more wins than Mark Price. Having Anthony Mason on your team, in other words, gets you one more win per year than having Mark Price does. That’s not insignificant.
But what if you’re last year’s Sixers, and you’re one shooter away from being really interesting? Price could take you to the next level… Mason will just relegate Marreese Speights, a pretty good young player, to the end of the bench. Anthony Mason might be one win better in a vacuum, but Mark Price, on your team, could be five or ten wins better. He could make up the difference in value between the two of them in a single season. Of course you’d take Mark Price over Anthony Mason.
I am by no means advocating that you should pass up a potential superstar to fill a specific little hole on your team. If there is a guy out there with more star potential than anybody else, you take him, no matter what position he plays. But often you’re not picking between Todd Fuller and Kobe Bryant. Often you’re picking between J.J. Redick and Hilton Armstrong, or Shareef Abdur-Rahim and Stephon Marbury. Quite often, you’re picking between two guys whose star potential is about indistinguishably similar, or two guys whose star potential is zero. In those cases, need can and should play a tie-breaking role. It is by no means as important as talent, but it is much more quantifiable, and there’s value in that.
by onlxn on Jun 24, 2009 11:46 AM PDT up reply actions 0 recs
Following a pure "BPA" strategy with the Hawks’ pick, at the time, would have probably led you — and did lead them — to Marvin Williams.
You’re confusing the strategy with the result. The smart bet was and is still to target the BPA.
Following a tiering strategy would have led you to Deron or CP3. The evaluations of those two and Marvin Williams were close enough at the time that there was no real consensus as to who’d be the best player of the three. In that scenario, where it’s more or less a tie, need can and should come into play… the Hawks would’ve picked one of the points.
Owen — dog — I’m pretty you can see that this is only the case because the Hawks happened not have a star PG. I’m honestly a little surprised that you’re having such a hard time distinguishing between the way a decision turned out and the wisdom of the logic behind the decision. If the Hawks actually thought Marvin Williams was the better player, statistically speaking, they made the correct decision, regardless of how it turned out.
Saying "the Hawks should’ve known Paul was the BPA and taken him" is not an endorsement of one drafting philosophy over another… that’s an argument for seeing the future
I see your point here. In the case of Paul v. Marvin Williams (a case you chose, remember) it’s quite possible that there were no obvious or even subtle signs that Paul was going to blow up the way he did; and that even the most finely-honed BPA approach would have led to Williams. In plenty of other cases (Bowie v. Jordan, Fuller v. Kobe, Foyle v. McGrady, etc.) a BPA approach would most likely have averted a major drafting blunder. In all cases, the BPA approach was / is / will be the correct one, since it doesn’t arbitrarily narrow your pool of options (sorry if I’m sounding like a corrupt mp3 here…)
Nobody can project player rankings perfectly, not even Jerry West. What you can project quite well is how your team fits together.
Really? I would say that the pseudo-science of player “chemistry” is even iffier than the pseudo-science of player evaluation.
Whew. OK, man… I think I’ve finally run out of steam. I’ve made my point amply clear; ditto you. There may some fundamental philosophical divide at work here. I’ll ponder it on the way home and if I think of something enlightening to say I’ll chime in here again. If not, thanks for a lively debate.
Thing 1
by Sleepy Freud on Jun 24, 2009 4:46 PM PDT up reply actions 0 recs
You’re confusing the strategy with the result. The smart bet was and is still to target the BPA.
That’s what they did… they ended up with Marvin Williams.
Owen — dog — I’m pretty you can see that this is only the case because the Hawks happened not have a star PG.
THAT IS EXACTLY WHAT I’M SAYING. If the Hawks were like most teams, they had Williams and Paul rated about evenly. If two guys are even, TAKE THE GUY YOU NEED MORE.
I’m honestly a little surprised that you’re having such a hard time distinguishing between the way a decision turned out and the wisdom of the logic behind the decision. If the Hawks actually thought Marvin Williams was the better player, statistically speaking, they made the correct decision, regardless of how it turned out.
No, they didn’t.
Your logic is tautological — that whoever gets the best player “wins”. What if Chris Paul had become, instead of Basketball Jesus, a pretty good point guard, a very slightly worse player than Marvin Williams? Would the Hawks still be in better shape as a result of correctly identifying the best player?
No, they wouldn’t. They still need a point guard. You don’t just match your player rankings against another team’s… you send five guys out there who have to play together. You replace Marvin Williams with a pretty good PG, even one who’s a little worse than he is, and the Hawks would’ve been better off.
Replace Marvin Williams with a PG who’s a lot worse than Marvin Williams, of course, and you’re screwing yourself. You can’t take a clearly worse player solely because of need. That’s how we ended up with POB.
But that wasn’t the case with Atlanta. They rated both guys just about equally, by all accounts, but they decided to give the edge to Williams, the guy who didn’t fit their team. That was a mistake, and it would’ve been even if Chris Paul hadn’t blown up.
I see your point here. In the case of Paul v. Marvin Williams (a case you chose, remember) it’s quite possible that there were no obvious or even subtle signs that Paul was going to blow up the way he did; and that even the most finely-honed BPA approach would have led to Williams.
Exactly. If you go back, a majority of pundits saw Williams as the superior player, albeit by just a little bit. That was a ginyoowine BPA pick.
In plenty of other cases (Bowie v. Jordan, Fuller v. Kobe, Foyle v. McGrady, etc.) a BPA approach would most likely have averted a major drafting blunder.
Absolutely. BPA should be the primary guiding philosophy behind any pick, whenever possible.
In all cases, the BPA approach was / is / will be the correct one, since it doesn’t arbitrarily narrow your pool of options (sorry if I’m sounding like a corrupt mp3 here…)
You ain’t sounding nearly as redundant as me, so no worries…
The thing that we’re not going to agree upon is the idea that you can always analyze two players and know which one is better. You often can… you usually can, even. But sometimes you just can’t. A lot of picks that have been made in league history have been toss-ups — the draft room’s split on two players, so they just throw up their hands and pick one. GMs and scouts and columnists and fans can pretend otherwise, but really, there are cases where you simply lack the insight to make an informed choice. And in those cases, you should pick the guy you need more.
If you don’t acknowledge that such cases exist, then indeed, you should stick to BPA. But I think BPA’s more of a religion than a science. It should be adhered to until you’re within the margin of error, and then other considerations should come in.
Really? I would say that the pseudo-science of player "chemistry" is even iffier than the pseudo-science of player evaluation.
Not talking about chemistry… talking about positions. The Kings are going to draft a point guard tomorrow, because 1) there will be available points that strike them at least as good as the other available players, and 2) they really, really need a point guard. If James Harden falls to them, and they calculate that James Harden will be 0.0001742% better than the PG they like best, should they take Harden?
No. They shouldn’t. That gap is too small to be meaningful. And they have to field a basketball team.
If not, thanks for a lively debate.
And thank you.
by onlxn on Jun 24, 2009 5:09 PM PDT up reply actions 0 recs
OK, I gotta speak up here, partly ’cos I think I should be allowed the last word in one of these two exchanges, but also ’cos this is just beyond the pale…
Your logic is tautological — that whoever gets the best player "wins".
Wow, man. Precisely the opposite. You’re the one using the Paul v. Williams to discredit BPA because, in this instance, the Hawks “lost.” It’s really not a whole lot more “logical” than using the same draft to support the idea that you should always pick guys from Wake over guys from UNC. Given your history of eloquence and tight logic on this board, I’m a bit shocked.
Rest of your post makes more sense. I’ll allow that the added value of positional “fit” with your current team is greater than .0001742%. Leave it to jae to work out the precise number.
Thing 1
by Sleepy Freud on Jun 24, 2009 6:12 PM PDT up reply actions 0 recs
You do merit a last word, so feel free to respond to this.
I wasn’t sufficiently clear. Your contention is that picking the person you deem to be the best player, by no matter how small a margin, is always the right move. I don’t think that’s right, in the sense that that strategy can cost you wins and make your team worse.
Earlier in the thread, you asked this:
you point to even one situation in which picking the BPA proved, short term or long term, to be an unwise move?[/blockquote]In response, I brought up Williams over Paul, an example I stand by. But that’s not because Paul turned out better than Williams after the fact. It’s because, at the time, the Hawks had no compelling reason to think he was worse than Williams, and they had every reason to think he’d help them win more games in the near future than Williams. The fact that Paul turned out as good as he did made it sting a little more, sure, but that’s not what made it a mistake. It was a mistake on draft day, when the Hawks had two guys they liked just about equally, two guys with equal star potential — co-BPAs, if you will — and picked the guy who didn’t fit their roster.
That’s my point. BPA isn’t wrong… it just doesn’t account for “too close to call” situations. In those situations, the smartest thing to do is to look at fit.
Rest of your post makes more sense. I’ll allow that the added value of positional "fit" with your current team is greater than .0001742%. Leave it to jae to work out the precise number.
HOLLA! (Watch the Kings take Harden now and completely clown me.)
by onlxn on Jun 24, 2009 11:45 PM PDT up reply actions 0 recs
But that’s not because Paul turned out better than Williams after the fact. It’s because, at the time, the Hawks had no compelling reason to think he was worse than Williams, and they had every reason to think he’d help them win more games in the near future than Williams. The fact that Paul turned out as good as he did made it sting a little more, sure, but that’s not what made it a mistake. It was a mistake on draft day, when the Hawks had two guys they liked just about equally, two guys with equal star potential
Now we are trying to guess how they evaluated the talent on the board. We know that they valued Williams more than Paul, but no one can say how much more.
I remember at the time Williams was thought my many people to be in a tier above Paul and Deron. Those people were obviously wrong, but it was thought to be a reasonable assessment, and it could have very well been the opinion of the Hawks front office. We really don’t know.
Thing 2
by olympicmike on Jun 25, 2009 12:19 PM PDT up reply actions 0 recs
It’s true… we don’t. We know that Paul greatly impressed them in workouts (as did Deron)… we also know that they’d decided on Williams a bit before the draft, so it didn’t come down to the wire for them. (If you ask Hawks fans, the pick was made because Billy Knight is obsessed with guys who are 6’8" and 6’9", and he has called that “the perfect NBA height” in the past, but we shouldn’t regard that as proof of anything.)
The overall impression was indeed that Williams would be better than Paul and Deron, with maybe 2/3s of the various pundits favoring Williams over Paul. I don’t recall any talk of Williams being a real level above them, however… I remember a lot of talk about a “four-player draft”. (Funny, in retrospect, that Bogut went above all these guys… not that he’s terrible, but still.) Not that it should matter, but most Hawks fans, on blogs and elsewhere, were clamoring for a point guard, though admittedly some were advocating trading down to do it (which would’ve been smart). Taking Williams was by no means considered a no-brainer by everyone.
Look, I’m generally a believer in BPA. And if you want to claim that the Hawks were justified in picking Marvin Williams based on what they knew at the time, fair enough… I don’t have a problem with that. But again, Sleepy asked this:
you point to even one situation in which picking the BPA proved, short term or long term, to be an unwise move?[/blockquote]I don’t see how “Marvin Williams over Chris Paul” is NOT an answer to that question. The Hawks took the guy they thought was better over the guy that fit their team, a guy that was universally considered to have star potential himself, and it cost them a chance to dramatically upgrade their team. Saying “if they thought he was the BPA, it wasn’t a mistake” is to excuse any draft pick ever, is it not? Couldn’t Stu Inman simply say, “you can’t criticize me for taking Sam Bowie, I thought he was the best player available”?
GMs make bad picks all the time… it’s the nature of the beast. But I think it’s kosher to point out that a single-minded adherence to BPA has hurt teams before. That’s what happened to the Hawks in 2005.
by onlxn on Jun 25, 2009 1:30 PM PDT up reply actions 0 recs
Saying "if they thought he was the BPA, it wasn’t a mistake" is to excuse any draft pick ever, is it not? Couldn’t Stu Inman simply say, "you can’t criticize me for taking Sam Bowie, I thought he was the best player available"?
A mistake is still a mistake, but it doesn’t mean it was the wrong strategy. That’s the difference I’m seeing here. Obviously if you fail to identify the BPA (which happens all the time) you have made a mistake. You can’t excuse that by saying “i thought he was the BPA”. but making a poor judgment about a player doesn’t really do anything to discredit the strategy of selecting the BPA regardless of position.
I understand your point that sometimes there is such a small difference that you can use position as a tie breaker. I think most team owners would expect their GM’s to have strong opinions about the players that it’s their job to assess though. I would hope that all that scouting and evaluating would give them a clear picture (right or wrong) of who they think the best player available is in any given scenario.
GMs make bad picks all the time… it’s the nature of the beast. But I think it’s kosher to point out that a single-minded adherence to BPA has hurt teams before. That’s what happened to the Hawks in 2005.
That’s the thing though, “a single minded adherence to BPA” wasn’t what hurt them, it was not having their draft board in the right order. Williams is clearly nowhere near the player that Paul is, but if he was just as good nobody would care that they passed on CP3. I just don’t think that position factors in that much at that point. Yeah they needed a PG bad, but more than that they just needed a really good player. If they struck gold with Williams their PG problems would have been an afterthought. Do you think anyone would be mad about this line-up
PG: Bibby
SG: Johnson
SF: Williams (in this theoretical scenario a top 5-10 player in the league)
PF: Smith
C: Horford
If Williams turned out about as good as Paul, that could be a title contender. You can address needs later, when you get a free pick at a player you should just focus on getting the best guy. That in itself is hard enough.
Thing 2
by olympicmike on Jun 25, 2009 2:59 PM PDT up reply actions 0 recs
Williams is clearly nowhere near the player that Paul is, but if he was just as good nobody would care that they passed on CP3.
If he was that good, certainly you wouldn’t have had Hawks fans calling for Billy Knight’s head. Still, an as-good-as-CP3 Marvin Williams wouldn’t improve the Hawks, then or now, as much as CP3.
Do you think anyone would be mad about this line-up
PG: Bibby
SG: Johnson
SF: Williams (in this theoretical scenario a top 5-10 player in the league)
PF: Smith
C: Horford
If Williams turned out about as good as Paul, that could be a title contender.
That’d be a very good team, for sure. But that’s ignoring the 2 1/2 years they spent without Bibby. In Marvin Williams’s rookie year, the Hawks went 26-56… they were one of the worst teams in the league. Their biggest flaw? No point guard. Williams’s second year, they went 30-52, another thoroughly crappy performance. Biggest reason why? No point guard. Williams’s third year, they added Bibby, and limped to the eighth seed despite a 37-45 record. A little better, because they had a point guard part of the year. This year, they had a credible point guard all season and were a good team.
If you magically turned Marvin Williams into a top-ten player, all these records would have certainly improved… maybe they’d have slipped into the playoffs in his second year instead of his third. But they’d still have been missing a point guard, and still struggling at times because of it.
Add Chris Paul instead of this new, better Marvin Williams? Giving you a Paul/Johnson/Childress/Harrington/Pachulia starting lineup that first season, with Josh Smith off the bench? One never knows, but I’d guess that team would do a good bit better than Lue/Johnson/Superwilliams/Harrington/Pachulia, with Childress and Josh Smith off the bench. The latter team still would’ve needed a point… even great players profit from good points. Look at Amare, with and without Nash. Look at how much better the Cavs got when Mo Williams came along… he wasn’t the only reason their record skyrocketed this year, but he played a big role.
We’re probably not going to agree on this, and that’s fine… I totally get where you’re coming from. But I don’t accept the premise that if two players are equally good, it doesn’t matter which one you take. The Hawks really needed a supergreat player, but they also needed a point guard. Fit matters.
by onlxn on Jun 25, 2009 4:14 PM PDT up reply actions 0 recs
We’re probably not going to agree on this, and that’s fine… I totally get where you’re coming from. But I don’t accept the premise that if two players are equally good, it doesn’t matter which one you take. The Hawks really needed a supergreat player, but they also needed a point guard. Fit matters.
That’s not my premise. I just don’t think you come into the draft having two players rated the same if you’re a GM. If you really can’t distinguish between two guys, then sure take the guy that fits. I just think that there are probably other things that would separate them.
Thing 2
by olympicmike on Jun 25, 2009 4:57 PM PDT up reply actions 0 recs
In Marvin Williams’s rookie year, the Hawks went 26-56… they were one of the worst teams in the league. Their biggest flaw? No point guard. Williams’s second year, they went 30-52, another thoroughly crappy performance. Biggest reason why? No point guard.
That has absolutely everything to do with the Billy Knight and Hawks management for being unable to get anybody that can pass as a point guard. That has nothing to do with your BPA argument. Marvin Williams was projected to be a better player in the NBA than Williams and Paul,, the people who made the projections were wrong in the sense that they didnt see Williams and Paul playing for good coaches and developing into extremely dynamic players. Marvin Williams is not a bad player. Marvin Williams is good young solid player. You said that Williams might have led to loosing Childress down the line a few posts back, again this is Hawks management in action nothing to do with draft strategy.
Add Chris Paul instead of this new, better Marvin Williams?
Easy to say now, who thought Chris Paul was going to be Isiah Thomas 2.0. Nobody. Again, the Hawks made a decent draft pick, they just picked the wrong guy, One guy turned out to be a solid NBA player and the other will rewrite history for the pg posisition but there was no way you or anybody else saw that coming from college.
But I don’t accept the premise that if two players are equally good, it doesn’t matter which one you take.
Ok, just like affirmative action, if 2 people are completely equal, take the minority. When are 2 players absolutley 100% completely equal. Hill vs Curry, you can’t make that pick on need, you got to get someone to decide who will be a better player and who will help the team win more games.
Thing B
by warriorsscore110 on Jun 25, 2009 10:13 PM PDT up reply actions 0 recs
sleepy you hit it right on the head,
drafting BPA available is much different than drafting on potential. DeRozen and Holliday will be picked 1 and 2 if you are just shooting for chance that a kid will be the next great thing. As far as the Mark Price vs Anthony Mason comment, I [say we are a 29 win team. We suck lets just get more talent and have better players on the floor.
Thing B
by warriorsscore110 on Jun 25, 2009 9:52 PM PDT up reply actions 0 recs
Need vs. Best Possible Athlete
The W’s have been doing that for years, but the problem with the W’s is that we keep drafting Athletic but skinny bigs, like 6’10 185lbs or 6’7 PFs. Or athletic SG that are suppose to develop into PG ala Hughes, Arenas but never really become PG just small SGs; is Ellis really any different? You can say that assists were up last year without a true PG on the roster but what they fail to take into account is when it really matters in the waining seconds of a close game, how many times did we loose a possession on a turnover or ill-advised three pointer? That is when you need a PG not SJack dribbling till 3 seconds are on the clock.
I asked Bill Walsh once on KNBR about that and he said it doesn’t make sense to draft the best available athlete in positions you already have good players in because those guys never have a chance to develop. He prefered drafting the best possible athletes in the positions of need. Forget Hill, we already have Wright and Randolph, give them a chance to grow and gel. At SF and SG we have a log jam with Maggette, Morrow, Jackson, Belinelli, Azubuike and throw in Ellis. I am confident in our youth, give them a chance before breaking them up again. We did that with the Webber team, the Jamison team, the Murphy/JR/Arenas team and recently the We Believe team, breaking up the teams or drafting the next redundant player. We have the peices, just close the deal and get us what we want, what the team needs, a PG not another PF to rott on the bench like Wright and Randolph. Look what happened when you gave them PT.
by Ultimate Warrior on Jun 25, 2009 4:08 PM PDT up reply actions 0 recs
ok,
There is a big difference between best player available and best possible athlete.
You can say that assists were up last year without a true PG on the roster but what they fail to take into account is when it really matters in the waining seconds of a close game
two things,
1, assists are not a sure fire mesurement of how good a point guard is or if he is a true point guard
2, idealy a good point guard will put your team in a situation where we don’t need Monta Ellis to hit game winning jump shots.
I asked Bill Walsh once on KNBR about that and he said it doesn’t make sense to draft the best available athlete in positions you already have good players in because those guys never have a chance to develop
fair enough, but football is a 3-5 year sport. You never know where the team will be at any posisition 5 years from now. In the NBA, you can and do wait atleast a couple of years for young players to develop. I generally judge a player at their 3rd year to see how good a player will be.
Forget Hill, we already have Wright and Randolph
Forget Wright, not that I don’t like the kid but no point in having him if Nellie wont play him. He is young and we can get something for him or package him with some of our other garbage.
I am confident in our youth, give them a chance before breaking them up again.
Not me pal, we won 29 games last year. We suck. Never make a trade for the sake of making a trade but Riley better be on the phone 10 hours a day trying to see what deals our out there.
Thing B
by warriorsscore110 on Jun 25, 2009 10:24 PM PDT up reply actions 0 recs
I asked Bill Walsh once on KNBR about that and he said it doesn’t make sense to draft the best available athlete in positions you already have good players in because those guys never have a chance to develop.
And in football, where the roster changes rapidly and where teams regularly rely on rookies for important roles, this may be true. Basketball, with a 12 man active roster and football, with 40-something players on a squad are different games. Down by one late in the game with a man on first, no outs and the pitcher at the plate, it’s a good idea to bunt. That’s about as relevant.
by jae on Jun 25, 2009 10:54 PM PDT up reply actions 0 recs
Different but the same
How many 40 somethings are on the W’s roster? How many times have the players touted as cornerstones to the franchise been turned over before an identity or chemistry could be formed? I don’t think that theary holds water relative to the W’[[COMMENT_CHILDREN_TOKEN]]#8217;s.
The games may be different but the management and motivation of resources, mainly personnel is the same. If you look at the true dynasties, the Bostons, San Antonios of the world, The 49ers,Steelers etc. Consistency is the core to success. I don’t care what sport it is, if you keep the core players and bring in youth from the draft to learn and eventually become the successors, while carrying the core values and tradition of that franchise, an identity and culture of winning is passes and perpetuated.
Unfortunately for the W’s, like the Clippers, we are the opposite, we draft guys, bury them on the bench with limited PT, and before they can develop, draft, trade or sign a similar player to further prevent any semblence of PT for development, only to jettison that player before they come into their own and we can actually get behind them. That has been our MO, a vicious cycle. Unfortunately in the last sixteen years, Nellie and his regime has been instrumental in that identity.
by Ultimate Warrior on Jun 30, 2009 1:01 PM PDT up reply actions 0 recs
i think this is pretty much what most people who chant “BPA” are saying, myself included. my one quibble on this is that your tiers are huge. if your scouts look at 6 guys projected as middle lottery players and say, “meh, they’re all about the same,” fire your scouts. tiers should be about 2 or 3 players in the lottery and get bigger as you get farther from the top of the draft. for example, here’d be about where i’d tier guys (and i’ll admit to having a controversial ranking of players, but hey, i’m not a scout):
tier 1: griffin
tier 2: rubio
tier 3: harden
tier 4: curry, hill
tier 5: lawson, williams, flynn
tier 6: derozan, holiday, jennings
tier 7: clark, thabeet, evans, johnson (taking evans or thabeet in the top ten screams bust to me)
tier 8: henderson, maynor, claver, and probably a couple other guys i can’t think of
heart of a champion, will of the warrior.
by cap'n hack on Jun 24, 2009 12:21 AM PDT reply actions 0 recs
oh, and apologies, i just noticed that “your tiers” are actually chad ford’s. chad ford needs to narrow his tiers. as far as i know, your yet unrevealed draft borad is absolutely perfect.
heart of a champion, will of the warrior.
by cap'n hack on Jun 24, 2009 12:28 AM PDT up reply actions 0 recs
EggZACKLY, dog. I ain’t wasting my flawless draft board on y’all.
I think we’re pretty much agreeing. Again, I can’t really rate/rank this year’s crop with much confidence… I stopped paying super-close attention to the college game when all the good players started leaving after one season. It’s very possible and even probable that Chad Ford’s tiers are overly conservative. He seems like the type who’d rather seem inoffensive than opinionated.
But I think the point still stands. Rubio has much more upside than Harden, but has contractual issues and health issues that at least partially close the gap. There are folks who’d rank Lawson as the 2nd-best player and folks who wouldn’t put him in the top twenty. A guy you’ve placed in your seventh tier is reportedly garnering interest in the top three.
I don’t say any of this to criticize your rankings. Just making the point that a draft can be — and, this year, seemingly is — a really murky thing. This statement…
tiers should be about 2 or 3 players in the lottery and get bigger as you get farther from the top of the draft.
…might hold true in most years, but I don’t know why it’d have to be accurate every season. Surely there are some years when the top talent clumps together more tightly than other years… your scouting department wouldn’t be helping you by creating arbitrary divisions whether or not they apply to the existing talent pool.
I think it’s useful for fans to recognize the differences between their incentives and those of NBA front offices. For us, it’s really fun to rank the best college & international players in a confident order, and to assert why we like Player X over Player Y. If you’re actually working for a team, though, I don’t think orders are so simply achieved. In the absence of consensus over superior talent, you may be drawn to the guy who fits into your organization the best. And I think that’s not only reasonable, but smart.
by onlxn on Jun 24, 2009 12:41 AM PDT up reply actions 0 recs
you’re right that the 2 or 3 guys per tier in the lotto is more of a guideline than a piece of absolute, infallible wisdom, but i think to place 6 guys in tier 3 (picks 5-10) just smacks of laziness on chad ford’s part. i’d say that if making a consensus list of tiers is what you’re after (and very well could be what ford was doing), then that might make sense, but you pay scouts for a reason and if my team just started drafting based on espn buzz and draft express boards, i’d be upset.
your scouts should develop strong opinions about guys, because it is very possible that at least 3 of the players in ford’s third tier will be busts. it’s your scouts job to figure out which of those players are likely to be busts and drop them on your board. consensus will usually get it about right, but the scouts need to figure out how to separate the players out and make the right call. otherwise, how could we possibly choose between jennings and flynn? they’d fit the same needs, but shouldn’t be ranked identically. i’d even propose that within tiers, teams use rankings to decide which of the players that fill similar needs is the best of the bunch. those rankings wouldn’t change anything if you were choosing between jordan hill or stephen curry (then it’s fit), but would matter if you were deciding on either james johnson or earl clark.
but again, i think we agree on the majority of this stuff and the post itself was very well thought out and argued.
heart of a champion, will of the warrior.
by cap'n hack on Jun 24, 2009 12:52 AM PDT up reply actions 0 recs
Yup
Again, I can’t really rate/rank this year’s crop with much confidence…
I think this is the sentiment most everyone has right now, which in turn is causing so much indecision and speculation. Personally, I’d like to see us trade down (or even out) and acquire some proven commodities. I would much rather surround some of the guys we have here now with a vet than bring in another young guy, whether he be for need or BPA. Given we have so many wing players, use it to parlay acquiring a ‘need’.
Chris Cohan- YOU'RE FIRED!
by bonbrillio on Jun 24, 2009 7:49 AM PDT up reply actions 0 recs
Just to more or less agree with cap’n hack – I’m a big BPA person, but, I think the tier system is the way to go. I also think a players “fit” not at position, but into team philosophy should factor into what tier they’re in. That is, someone slow that won’t do much fast breaking won’t fit well into our system. Even if they’re more talented than someone else, if they don’t have the skills Don Nelson likes they’re going to use that talent warming the bench. Other, less talented guys (Matt Barnes anyone?) will be able to contribute more, though, if they’re a good fit. So I see that as a factor in what tier a player is in. How much of a factor? For someone like Don Nelson it’s a pretty big considering he’s a bit…unorthodox, let’s say…but for many other coaches it will be a much smaller factor, so that’s something that each FO needs to determine.
by Missing Barry on Jun 24, 2009 7:43 AM PDT reply actions 0 recs
I have no problems with tiers. I do however believe that the moment that you let “need” start to creep into your talent evaluation, your tiers start to fall apart and you start to convince yourself that you really can’t tell the difference between Foyle and McGrady, the the POB really could be the 9th best player in the draft, we don’t need another PF, therefore these big point guards are really equal talents to the PF other people seem to regard as a talent. Scouts need to be able to make honest evaluations about what they think a player can be, and how likely that player will get to that level. Clouding that with “but we don’t need a , we really need a ” gets in the way of that. It seems to me that where it does the most damage is that it tends to make people ignore the probability of anyone realizing this mythological potential-beast in favor of the dream of what they’d like the player’s chances to be.
by jae on Jun 24, 2009 8:27 AM PDT reply actions 0 recs
I agree that “need” can be a dangerous bias when evaluating draft talent. But there are many biases that can affect talent evaluation, and need is far from the most pernicious. Off the top of my head, here are five evaluative biases that are more damaging than thoughts of need.
1) Biases based on who a player reminds people of. Insert your favorite Dunleavy-Bird comparison here.
2) Biases based on a player’s size, regardless of production. Most had Michael Olowokandi as that year’s BPA, not just the Clips.
3) Biases based on a player’s combine performance. Joey Graham SLAUGHTERED Monta at the combine.
4) Biases based on a player’s personality. Adonal’s considerable charm probably contributed to our overdraft of him.
5) Biases based on a player’s regional fanbase. AKA, why Felton and May are Bobcats and Danny Granger isn’t.
Need gets a bad rap sometimes. We Warriors fans are allergic to the idea of drafting for need because of a run of comically bad center picks. Fuller, Foyle, POB… we all know the list. But wasn’t Biedrins a “need” pick? We badly needed a center at the time, and nobody else had Biedrins that high… Biedrins was admittedly a project, but then so was POB. It seems to me that POB is denigrated as a “need” pick because he (unsurprisingly) turned out to suck, while Biedrins is regarded as a “BPA” pick because he turned out to be good. That distinction strikes me as arbitrary. I think that “need” was a factor in our selecting Andris Biedrins, and I’m pretty happy with how it turned out. Need combined with incompetent scouting gets you POB… need combined with good scouting can be just fine.
It seems to me that where it does the most damage is that it tends to make people ignore the probability of anyone realizing this mythological potential-beast in favor of the dream of what they’d like the player’s chances to be.
This is absolutely the biggest risk when drafting. You can not just fall in love with a guy’s best-case scenario… there are many more variables than that. How likely is he to reach his ceiling? If he doesn’t, how likely is he to be useful anyway? How soon will he be useful — will you get real value from him while he’s cheap and under your control? How healthy is he? How old is he? All sorts of biases get in the way. And need, in particular, has led this franchise astray several times in the past fifteen years. But our history is not actually that common… need is rarely the main variable that leads an evaluator astray. Usually it’s irrational optimism or a huge physique or just some constellation of elements that makes a drafter see a superstar where there isn’t one. (And I’d argue that even in our case, incompetence has been as much of an issue as need. Fuller over Kobe is obviously a nightmare, but we didn’t just pass up Kobe… we passed up Jermaine O’Neal and Zydrunas Ilgauskas and Vitaly Potapenko. It wasn’t just that we took a center over the best available player… we took the fourth-best available center over the best available player. Replace Fuller with Jermaine O’Neal and that draft wouldn’t sting quite as badly.)
Need can lead people to make stupid picks. But it’s not usually the culprit behind a stupid pick. Usually the culprit is… well… stupidity.
by onlxn on Jun 24, 2009 10:46 AM PDT up reply actions 0 recs
Usually the culprit is… well… stupidity.
Right, but the inability to grasp basic probability (in games of chance, you should never minimize your chances of success) and the inability to think long term (your so-called “needs” can change in the blink of an eye) are two of the hallmarks of stupidity. Or at least, less-than-jae-level intelligence. ;-)
Thing 1
by Sleepy Freud on Jun 24, 2009 11:21 AM PDT up reply actions 0 recs
But this is no regular game of chance
Some of these GMs have to roll a hard six, or else they’re out of a job. Player A is low risk, medium reward, and Player B is huge risk, higher reward because he fits a need and has more Potential; Player A’s max potential, OTH, doesn’t spell Top 4 Seed in the next two years. Sometimes they have to go with Player B and hope for the best. GMs might be able to think long term, but can’t always ACT on it because of the nature of their profession.
by ffgolden on Jun 24, 2009 11:46 AM PDT up reply actions 0 recs
the inability to grasp basic probability (in games of chance, you should never minimize your chances of success)
This is assuming that an evaluator can confidently order players in terms of chances of success. No evaluator in the world can do that accurately. You could ask every NBA pundit to rank who the top 20 players from this draft will be, and every single one of them will have gotten several things wrong in ten years’ time.
That’s not to say that you stop trying. You have to evaluate player potential as best you can… some people are better at it than others. And this…
the inability to think long term (your so-called "needs" can change in the blink of an eye)
…is certainly true.
But there are points — about a dozen NBA teams are currently at this point — where you will have trouble finding a meaningful difference in expected value between a couple players you’re looking at. You look at Stephen Curry and Jordan Hill, their ceilings, their chances of reaching those ceilings, their athleticism, their psychological profiles, their bodies, their favorite candies… and there is just no really strong reason to put one over the other. It’s a push. Fans might argue that there should never be a situation like this — that you should always be able to rationally favor one over the other — but that’s often just not the case. There’s too much guesswork in player projection for choices to always be clear. Some dumb variable will break the tie. GMs will draft one guy over another because his backstory is more easily marketed, or because he made more eye contact in his team interview than the other guy. Sometimes tie-breaking picks come down to the whim of an individual.
There’s a better way to break those ties: by thinking about the composition of your team. Not because that’s critically important, but because it’s a certainty. When choosing between two comparable commodities, disregard your gut and choose the commodity that fits your needs better. That’s just simple intelligence.
by onlxn on Jun 24, 2009 12:03 PM PDT up reply actions 0 recs
This is assuming that an evaluator can confidently order players in terms of chances of success. No evaluator in the world can do that accurately.
One can increase one’s chances of success without having to evaluate them accurately. Most smart people are pretty good as this. Smart Warriors fans say things like, “if we finished with the worst record, we’d have the best chances of getting the number one pick” and “if we got the #1 pick, the guy we picked there would have a really good shot of being a really good player.” Less smart Warriors fans say things (on this board) like “it’s all a crapshoot” and “Kwame Brown: I rest my case.”
On topic: limiting yourself to one or two positions of perceived short-term need, rather than drawing from the total available pool of talent, is lowering your chances of success. What the actual chances are now or were before is immaterial.
You look at Stephen Curry and Jordan Hill, their ceilings, their chances of reaching those ceilings, their athleticism, their psychological profiles, their bodies, their favorite candies… [snip]
I’m not talking about this specific case. If the Warriors brass, after factoring in the two players’ respective upsides, realistic chance of achieving that upside, and “positional values,” determines that they’re roughly equal prospects, decides to pick Hill, I have no issue with that. If they use “we’re thin at PF” or “we already have enough guards” as part of their calculus, I do have a problem with it.
Thing 1
by Sleepy Freud on Jun 24, 2009 12:44 PM PDT up reply actions 0 recs
One can increase one’s chances of success without having to evaluate them accurately. Most smart people are pretty good as this. Smart Warriors fans say things like, "if we finished with the worst record, we’d have the best chances of getting the number one pick" and "if we got the #1 pick, the guy we picked there would have a really good shot of being a really good player." Less smart Warriors fans say things (on this board) like "it’s all a crapshoot" and "Kwame Brown: I rest my case."
You seem to have borrowed my straw man… I’m not advocating throwing evaluation out the window. I’m advocating an awareness of the limits of evaluation.
Picking a player is NOT comparable to getting a draft slot. The latter features a clear, concrete hierarchy… the former doesn’t. You try to create a hierarchy as best you can, and some things (Griffin up top) are clear enough, but there will be points when you — me, Larry Riley, Sam Presti, Chad Ford, anybody — don’t really have a strong reason for slotting one guy over another. At that point, you have two options: you can randomly pick one guy over the other, or you can break the tie by looking how each player would fit into your overall picture. I side with the latter.
On topic: limiting yourself to one or two positions of perceived short-term need, rather than drawing from the total available pool of talent, is lowering your chances of success. What the actual chances are now or were before is immaterial.
Nobody’s arguing for that. Always take the best guy in the pool… if that guy exists.
If the Warriors brass, after factoring in the two players’ respective upsides, realistic chance of achieving that upside, and "positional values," determines that they’re roughly equal prospects, decides to pick Hill, I have no issue with that. If they use "we’re thin at PF" or "we already have enough guards" as part of their calculus, I do have a problem with it.
So you’re fine with the Warriors choosing Jordan Hill over an equally-promising player for no particular reason, but you’re not fine with the Warriors choosing Jordan Hill over an equally-promising player because of basketball considerations? You’d rather see the team flip a coin than give even a cursory thought to the team actually looks like? I’m sorry, dog — that strikes me as completely nonsensical.
by onlxn on Jun 24, 2009 1:03 PM PDT up reply actions 0 recs
You seem to have borrowed my straw man…
How so? You said no evaluator can accurately order players in terms of success. I said even if you can’t order them “accurately,” you can do to so maximize your chances of success — specifically, by not ranking them based on your perceived short-term “need.”
I’m not advocating throwing evaluation out the window.
And I never suggested you were. (I’ll refrain from invoking that anthropomorphized bale of hay, as he has a way of ruining good debates…)
I’m advocating an awareness of the limits of evaluation.
I understand that. And I guess I’ll reiterate, even though I think you get my gist: even if you accept that human methods of evaluation are highly iffy and chancy, there are still reliable ways to maximize (or minimize) your chances of success.
Picking a player is NOT comparable to getting a draft slot. The latter features a clear, concrete hierarchy… the former doesn’t.
Ain’t nothing particularly “clear and concrete” about a higher v. a lower pick (see Bowie v. Jordan, etc.). Your chances of success improve the higher you pick, same way your chances of success improve when you open your pool to all players, regardless of position.
So you’re fine with the Warriors choosing Jordan Hill over an equally-promising player for no particular reason,
In theory, no, I don’t have an issue with it. People make decisions based on roughly 50/50 odds all the time. If someone flips a coin and asks you to call it, you call it. In practice, I think true 50/50 bets are pretty rare. There’s usually a decisive factor (intangibles, gut feeling, etc.) that tips the balance — one that should outweigh perceived short term “need,” imo.
you’re not fine with the Warriors choosing Jordan Hill over an equally-promising player because of basketball considerations?
“Basketball considerations” is your term. Obviously I’m OK with using that as a criteria. Perceived short-term “need,” not so much.
Since you keep returning to the specific case of Hill v. Curry (probably wisely): I would say I think (1) Curry has a bigger upside, (2) Curry has a more realistic chance of achieving that upside; and (3) for the forseeable future, I’m more comfortable with Randolph/Wright manning the 4 than Monta/CJ/Law/Claxton manning the point. Obviously, #3 is the last of my considerations, but since all three are aligned in Curry’s favor, it’s kind of a moot point.
I’m sorry, dog — that strikes me as completely nonsensical.
Woof. ;-)
Thing 1
by Sleepy Freud on Jun 24, 2009 4:01 PM PDT up reply actions 0 recs
Ain’t nothing particularly "clear and concrete" about a higher v. a lower pick (see Bowie v. Jordan, etc.). Your chances of success improve the higher you pick, same way your chances of success improve when you open your pool to all players, regardless of position.
You misunderstood me. I said draft slot. The #7 pick, by mathematical definition, gives you more options than the #8 pick. 100 draft experts out of 100 would see more “upside” in the #7 slot than the #8 slot. But you’re equating favoring the #7 pick to the #8 pick to favoring one player over another. That analogy is contingent upon there being an objective ordering of players that you can determine. I think you can do that accurately in many cases — Griffin at one, etc. — but in many, you can’t… the variables are too complicated to produce a clear answer.
My pool is as open as yours. I’m just acknowledging that there is such thing as “too close to call”, and evaluators in many circles, professional and bloggy, do themselves a disservice by pretending otherwise. If you can not identify the best available player with any confidence, you can not base your decision solely on that metric. You have to take someone who you deem as good as everyone else available… if you’re still left with more than one guy, team fit’s the only tie-breaker that makes sense.
In theory, no, I don’t have an issue with it. People make decisions based on roughly 50/50 odds all the time. If someone flips a coin and asks you to call it, you call it. In practice, I think true 50/50 bets are pretty rare. There’s usually a decisive factor (intangibles, gut feeling, etc.) that tips the balance — one that should outweigh perceived short term "need," imo.
If by “intangibles” you mean the player’s attitude, intelligence, willpower, etc, that stuff shouldn’t be a tie-breaker… that stuff should be involved in the evaluation of the players already.
“Gut feeling” is exactly what I’m railing against. Good NBA general managers, scouts, etc., know exactly why they like a guy… they can explain what attributes put a player high on their list. But if a GM takes a guy based on “gut feeling”, some sort of instinct he can’t explain or analyze, he’s probably going to make a stupid pick.
Worrying about who your backup power forward is not a very good reason to pick one guy over another. But at least it’s a reason. “Gut feeling” is just a machofied excuse for laziness. Let’s not draft like George W. Bush, here.
"Basketball considerations" is your term. Obviously I’m OK with using that as a criteria. Perceived short-term "need," not so much.
I find the qualifier “perceived” to be strange. You think a general manager should be able to create a confident order of dozens of basketball players from all over the world, dozens of whom he’s never met, but you don’t trust him to have a sense of what his basketball team needs? You may not think a GM should draft based on need, but I think it’s fair to assume most GMs know what they need. That’s the easy part.
Are need considerations short-term? Sure, but so is everything about drafting.
Since you keep returning to the specific case of Hill v. Curry (probably wisely): I would say I think (1) Curry has a bigger upside, (2) Curry has a more realistic chance of achieving that upside; and (3) for the forseeable future, I’m more comfortable with Randolph/Wright manning the 4 than Monta/CJ/Law/Claxton manning the point. Obviously, #3 is the last of my considerations, but since all three are aligned in Curry’s favor, it’s kind of a moot point.
I’ve mainly been returning to it to keep this from getting too general… it’s more of a convenient stand-in than a prime example. But I agree with your analysis. (Actually, in this case, “need” doesn’t factor in much at all to me… our roster doesn’t have clear enough strengths or weaknesses anywhere to present a compelling case for any type of player. But, as you say, Curry seems far enough above Hill to me that positional considerations wouldn’t matter anyway.)
Woof. ;-)
SNARL
by onlxn on Jun 24, 2009 4:48 PM PDT up reply actions 0 recs
If your need is a big man, you should never consider need as part of the pick. The short supply of tall people suggests that this strategy will fail. If you’re lucky to have one at the top of your draft board when your pick comes up, go for it, but otherwise, you’re gonna get a stiff.
I do not recall many saying that Olowokandi was the consensus best. I remember there being more uncertainty that he’d get taken first than in many other years and that the Clippers inevitably thought that when you had a shot at a “franchise bigman” you had to take him. He wasn’t. Was it need? Was it just an error of scouting? Who knows.
Need can lead people to make stupid picks. But it’s not usually the culprit behind a stupid pick. Usually the culprit is… well… stupidity.
Clearly this is at the level of opinion, but I think that the other cases you presented often fall into the “need” category. Falling in love with a body type? Tends to happen when you perceive that body type as being your “need”.
Regional appeal doesn’t seem to fit into this, but that seems seldom to be an issue. There have been some terrible cases of it (Felton and May, previously J.R. Reid sticks out as well; hmmm: pattern of Charlotte franchises falling in love with former Tar Heels? Are there other clear cases?) but I’m struggling to think of many more. I can see far more cases where the need for a position caused a bad choice.
by jae on Jun 24, 2009 11:48 AM PDT up reply actions 0 recs
Clearly this is at the level of opinion, but I think that the other cases you presented often fall into the "need" category. Falling in love with a body type? Tends to happen when you perceive that body type as being your "need".
I would very much agree… “need” covers a lot of ground. We’ve actually seen a number of need-based arguments for Jordan Hill lately — the theory is that as a guy with a jump shot, he’s a more Nellie-friendly big than Biedrins or Wright.
I may just have less faith in the idea of a knowable order of player values than other people here. The broad strokes work just fine — Blake Griffin is very clearly a better prospect than anyone else, and should be taken ahead of them as a result. Danny Granger was a clear cut above the other talents available in the teens four years ago. I would never advocate taking a “need” guy over someone you KNOW will be a better player.
But you often just don’t know. The differences in expected values of players can often be tiny, well within any margin of error you’d estimate. At that point, you’re relying simply on a coin flip, and I don’t like that… that’s irrational drafting. When a choice between two players is too close to call, it’s better to try to address your needs than to simply flip a coin.
by onlxn on Jun 24, 2009 12:12 PM PDT up reply actions 0 recs
ayee
We USED to Believe...
WE DEMAND IMPROVEMENT!
by RunNdGun on Jun 24, 2009 4:00 PM PDT reply actions 0 recs
what any of us think doesn't matter
what Nelson thinks, or rather does in the draft matters:
http://blogs.mercurynews.com/warriors/2009/06/17/nelson-on-draft-day-an-appreciation/
by hardcore on Jun 25, 2009 1:55 PM PDT reply actions 0 recs
If you have two players that you perceive to have an identical ceiling, chance of hitting that ceiling, floor, and chance of hitting that floor
You choose the one at a position you “need.” It’s like the coin toss of the NFL’s method of determining which team with the identical record goes to the playoffs.
However, just like you rarely get to the coin toss in the NFL, that rarely, if ever, happens.
BPA is the best strategy. You have a better likelyhood of having a better player. Better player means more talent, more talent means more wins or better trade possibilities. The problem comes in the inherent uncertainty of the draft, and the inherent stupidity of GMs.
Just like you can throw a perfect change up knowing that the batter is sitting on a fastball, only to have him knock in a end of the bat chopper that turns into an infield hit.
Just like you can draw up a great prevent defense, only to knock the ball to Franco Harris and have him score a touchdown (when he completely dogged it earlier on in the play).
Just like you can put your opponent in poker all in with only one card in the deck that can help him and lose.
Just like any strategy can backfire due to unrelated, unpredictable circumstances, BPA can easily fail. The fact that it fails sometimes or in specific circumstances should not be used to suggest that the strategy is a failure.
In rare cases, two players can be so close that you should pick for need. In rare cases, two NFL teams will need to flip a coin to determine who goes to the playoffs and who stays home.
You have been DFiBrillated.
by Dubs fan in Boston on Jun 30, 2009 11:56 AM PDT reply actions 0 recs
Comments For This Post Are Closed
Secondary Sidebar
User Tools
FanPosts
Community blog posts and discussion.
Certain photos copyright © 2010 by Associated Press or Getty Images. Any commercial use or distribution without the express written consent of Associated Press and Getty Images is strictly prohibited.
