Warriors 2014-15 win projections with and without Kevin Love

Let's recap where I stand in terms of understanding the Kevin Love situation:

1. The Cavs have loosened their stance on trading Wiggins.
2. LeBron James is in Kevin Love's ear like Lance Stephenson was in LeBron James' ear.
3. The Warriors front office Voltron of minds still seems to be leaning towards not trading Klay Thompson (note: I almost went with Power Rangers metaphor there).

On the face of it, #3 seems outrageous. Kevin Love is the stretchiest of stretch 4's in the game and an elite rebounder. The Warriors actually don't need so much help rebounding, and Love's defense is not demonstrably better than David Lee, but his Reed Richards-like ability to stretch the floor make Love an elite commodity in the NBA and would potentially open up the floor for Curry, Iguodala, and even Barnes Shaun Livingston to slash to the basket more often. One only need to look at the free agency ongoings this summer to see where the priorities of front offices around the league are focused, as the likes of Josh McRoberts, Channing Frye, Patrick Patterson, Anthony Tolliver, and Marvin Williams demonstrate the increasing value of the stretch 4 (or 5) in the NBA. Heck, Steve Kerr kicked off his Golden State tenure by proclaiming that he'd like to see the Warriors go after a stretch 4. Heck, imagine the increase in value of David Lee if he could just hit 3pt shots (apparently he's trying real hard too!).

So it seems it's a no-brainer that the Warriors should trade Klay and Lee for Love...and most likely Kevin Martin. I say seems, because I haven't actually seen many (any?) bloggers or fans actually put a number on it. How many more wins can we reasonably expect with Love? Well, one fairly simple and unbiased way to make a win projection is to use one of the adjusted +/- metrics available. The one I currently trust the most is ESPN's "Real +/-" or RPM developed by Steve Ilardi and Jeremias Engelmann. These are two of the brightest NBA stats guys I know of, but more importantly than justification by authority, is the fact that their metric was developed to satisfy one master: predictive power. That, after all, is what we're after.

Let's start with the case where no trade happens. I first had to project minutes for each player. I played it pretty conservative here, using last year's minutes in most cases, except bumping up Draymond quite a bit, bumping Barnes down, and taking a couple minutes away from Lee. Because RPM is a possession-based stat, minutes must be converted to possessions, and combining the possessions and the RPM stats (from ESPN) with a formula for WARP (wins above replacement player), I get a projection of roughly 55 1/2 wins next season. This does not account for improvement or decline of players, injuries, or coaching. Of course, those factors would weigh in either scenario.

Keep Klay

https://docs.google.com/spreadsheets/d/1JoSkLvlRPEox8gQdDG-lqqHiUT1zyqlgakH8FU6W1hE/pubhtml

Trade Klay and Lee

In the case that we trade Klay and Lee, I assume we would take back Kevin Martin. I also assume that Livingston would become the starting SG and Martin would be the primary backup. Now, we can argue all day whether this is actually how it would go down, but I would point out that most of the alternative scenarios you can propose with this set of players won't lead to a very significant difference in win totals (I'll tell you how you can try in a little bit). The bottom line here is that the projected win total using RPM and these minutes is just about 58 wins, resulting in +2.3 wins compared to the Love-less scenario.

https://docs.google.com/spreadsheets/d/1GAxSCGBshL10-wYEe41diSO73oyNTFvUTXxnENdgkC4/pubhtml

Now, you certainly don't have to agree with the way I distributed minutes. I encourage you to download the spreadsheets and enter your own numbers. You can also use the spreadsheet as a template for any other teams you like (just look up RPM on the ESPN site). I think you'll find that even if you try very hard, you won't be able to easily come up with wildly different win totals than I show here, or I should say, very different totals for the scenarios with and without Love.

What is suggested by this set of projections is that although Love is a sizable upgrade from David Lee (+ 8 wins), the downgrade from Klay to Livingston/Martin is also sizable and negates much of that upgrade. Clearly, in the event that this trade were made, it would behoove the Warriors to find an upgrade to the SG position, although that might be very difficult given their cap situation going forward, especially given that Love is due for a massive new contract in a couple seasons. I do think there is some method to the madness of signing Klay to a max deal before the salary cap is raised significantly in a couple of summers. And we have seen that play out now with Hayward and Parsons. If Klay can be signed on a 5-year deal, say, for $80M ($16M) next summer, while it might sound "crazy" right now, it may be seen as a relative bargain in just a couple seasons (much as Curry's "risky" deal a couple seasons ago is looking like a massive bargain now).

To summarize my thoughts. I know most people think this trade is a no-brainer. I don't quite agree with that. When you take into account everything involved (and I haven't even come close to doing that myself), I think there's a little bit of brain needed. I think I still lean towards doing the deal. But I certainly understand why the FO might be thinking the way they are. There are multiple levels of decision-making that need to be made here and I hope what I've shared here starts to make that more clear.