A Win Shares Projection for Next Season

[PhoSita]

Many people on this board have argued that if the roster is kept the same, we could be looking at a team that wins 35-45 games and competes in the first round of the playoffs.

Others think this team is headed for a dive in the standings.


I'm aware that Win Shares and WS/48 are flawed stats, but I thought it would be interesting to use them to get a sense of what we could expect from this team if there are no major moves.  That said, I fully expect Ainge to do some more house cleaning and deck shuffling.  The results of this exercise demonstrate why that's necessary, in fact.


A note on the estimations:

For WS/48 I did a rough average based on the last few seasons for each player.  I did kind of a guesstimate as to how many games each player would appear in, and what their minutes would be like, but I think it's a reasonable guess.  Guys will probably sit out for even minor injuries, but I don't expect the team to blatantly hold people out.

The number in parentheses is WS/48.

Rondo - 52 games, 36 minutes per game (.12) = 4.68
Bradley - 72 games, 28 minutes per game (.0095) = .399
Green - 77 games, 36 minutes per game (.095) = 5.49
Sullinger - 62 games, 28 minutes per game (.14) = 5.06
Bass - 82 games, 24 minutes per game (.12) = 4.92
Wallace - 67 games, 18 minutes per game (.1) = 2.51
Humphries - 67 games, 18 minutes per game (.115) = 2.9
Lee - 77 games, 24 minutes per game  (.08) = 3.08

Total: 29.03 wins


This obviously doesn't include the end of bench guys, or Kelly Olynyk.  How many more wins could we expect to get from them, though?  Maybe 3 or 4 at most, is my guess.  In any case, we're looking at a team that's probably solidly on the upper end of the 25-35 wins range. 

In other words: the team as constructed would be bad, but not atrocious.  They'd be safely outside of the playoff hunt, but not bad enough to truly be scraping the bottom of the standings.  I think it's safe to say they're more than one or two pieces (Josh Smith / Big Al) away from being a respectable playoff squad, though.

Verdict: Ainge has some more cleaning to do, without a doubt.  Of course, the new coach could just not play Wallace or Humphries at all and give Lee and Bass much more limited minutes in order to let the youngsters get more time.  I don't think that force feeding younger players minutes at the expense of veterans is necessarily a good idea, though.  You want guys to feel like they have to actually work for and earn their minutes.

[RLewis35]

I appreciate the thoughtful stats based analysis but I think your post demonstrates just how flawed advanced metrics can be.  This shows that sullinger  bass are all more or equally valuable to rondo per 48 and humphries green and  Wallace are close to as valuable.  We all know that's absurd.  I think a Rondo led team w athletes will result in 40ish wins no matter what.  We will see (presuming rondo is on the team and plays all year).

[Who]

I pretty much agree with that forecast. I have the team pegged somewhere around 30-35 wins right now. Maybe a little less or little more depending on how much time Rondo misses.

I think a low level star like Big Al or Josh Smith can get the team up to 42-46 win territory and 6th-7th seed. But I think even a non-star, a DeAndre Jordan type, could be enough to get them to .500 and into the playoffs.

[PhoSita]

I appreciate the thoughtful stats based analysis but I think your post demonstrates just how flawed advanced metrics can be.  This shows that sullinger  bass are all more or equally valuable to rondo per 48 and humphries green and  Wallace are close to as valuable.  We all know that's absurd.  I think a Rondo led team w athletes will result in 40ish wins no matter what.  We will see (presuming rondo is on the team and plays all year).

It's true that Win Shares is a flawed stat for that reason, although I think an argument could be made that perhaps it doesn't underestimate Rondo's value THAT much when you consider the fact that the Celtics were not really any better with him last year than without him.

There's something to the idea, I think, that Rondo can really help a team that already has a lot of talent but probably isn't a great fit for a team where he's the best player -- at least in the regular season.

Anyways, Win Shares assumes that big men tend to be more valuable because they grab rebounds and score more points per shot, generally speaking.  I think that's probably true, most of the time. 

The thing with Rondo is that he's the sort of player who will make other players more valuable per minute, rather than generate value himself.  So if it helps just take a third of the wins from the big men and add it onto Rondo's number.

I think you could say that however much the stat undervalues Rondo, it proportionately overvalues guys like Sullinger, Bass, and Humphries.

The proof is in the pudding, so to speak -- I think the wins projection looks pretty spot on.

[rondoallaturca]

I also question the validity of these statistics, but the end result actually seems fairly accurate. When comparing our team as it is with the rest of the league, and particularly the East, we'll be looking at around 30 wins right now.

[Boris Badenov]

Nice analysis, TP. I was hoping to do something similar but you beat me to it.

I think this might be on the high end, for what that's worth. Ainge and our new coach can probably do many things to make the team worse (like trade Rondo), but probably will not do too many things to make the team better.

[BballTim]

Many people on this board have argued that if the roster is kept the same, we could be looking at a team that wins 35-45 games and competes in the first round of the playoffs.

Others think this team is headed for a dive in the standings.


I'm aware that Win Shares and WS/48 are flawed stats, but I thought it would be interesting to use them to get a sense of what we could expect from this team if there are no major moves.  That said, I fully expect Ainge to do some more house cleaning and deck shuffling.  The results of this exercise demonstrate why that's necessary, in fact.


A note on the estimations:

For WS/48 I did a rough average based on the last few seasons for each player.  I did kind of a guesstimate as to how many games each player would appear in, and what their minutes would be like, but I think it's a reasonable guess.  Guys will probably sit out for even minor injuries, but I don't expect the team to blatantly hold people out.

The number in parentheses is WS/48.

Rondo - 52 games, 36 minutes per game (.12) = 4.68
Bradley - 72 games, 28 minutes per game (.0095) = .399
Green - 77 games, 36 minutes per game (.095) = 5.49
Sullinger - 62 games, 28 minutes per game (.14) = 5.06
Bass - 82 games, 24 minutes per game (.12) = 4.92
Wallace - 67 games, 18 minutes per game (.1) = 2.51
Humphries - 67 games, 18 minutes per game (.115) = 2.9
Lee - 77 games, 24 minutes per game  (.08) = 3.08

Total: 29.03 wins


This obviously doesn't include the end of bench guys, or Kelly Olynyk.  How many more wins could we expect to get from them, though?  Maybe 3 or 4 at most, is my guess.  In any case, we're looking at a team that's probably solidly on the upper end of the 25-35 wins range. 

  Briefly, I'd expect Rondo to play more than that and I'd expect his WS number to go up without PP and KG, I think you'll see a higher number than that from Green, I think either Bradley or another sg will put up better numbers than that and you've left a good 1/4 of our minutes unaccounted for.

[celtics2]

Rondo if fit will get bored getting the ball to players that don't know what to do with it and the bottom will fall out of the team. Unless someone steps forward and challenges for Player of the Year Award. My guess is our Players will be wearing small size jock straps. 35 wins.