50 Wins in Minnesota?

Bill Simmons was working his trade machine magic again today and thinks he came up with two trades that will get Minnesota to 50 wins next season.  In case you hadn’t noticed, the Wolves only won 17 games last year, so this would be quite a turn-around.  Maybe Bill is just off his rocker?  Let’s take a look.

Here’s the current Wolves roster: Beasley, Ellington, Flynn, Hayward, Johnson, Love, Milicic, Pekovic, Randolph, Ridnour, Telfair, Tolliver, and Webster.   Here are Bill’s suggested trades:  The Wolves would send out Beasley and Webster for Gordon and Daye from the Pistons, and Flynn and Pekovic (along with the pick they used tonight for Derrick Williams) for Gortat and Dudley from the Suns.  The question is, how much will these players help?  Enough to jump 33 wins?  I’m going to gather an array of metrics to see what the numbers looks like.

The chart above has the Wins Produced, ezPM100, RAPM, and APM values for each player from this past season.  Remember that average is .1 for WP and 0 for the rest.  There are a lot of bad players here; WP says that Ridnour and Love were the only above-average players; the other metrics like Tolliver and Love (APM also gives a little credit to Beasley and Wes Johnson).  So it is probably no surprise that Minnesota was terrible.  With Bill’s trades, how would they look next year?

As you can see, the metrics all think that Dudley is above average and Gortat gets three out of four votes.  Gordon and Daye are below average, but on the other hand they aren’t as terrible as Flynn or Pekovic and are arguably about the same as Webster.  So how should this team do?  Let’s allocate some minutes according to Arturo’s regular season minute breakdown, assuming that the order is something like Bill’s suggested line-up.  The chart doesn’t have Ricky Rubio, but I’m going to guess he gets the second-most minutes played on the team since the Wolves have waited so long for him.  Here’s what I get (sorry for the change in names; the order is the same):

Wins Produced can just be summed up to get the expected number of wins while the other metrics are points produced over the course of the season.  To convert that to wins, I sum up for the team and divide by 82 to get team per-game point differential, then multiply by 2.54 and add 41 (incidentally, you can convert individual player points produced to wins above average by multiplying by 2.54 and dividing by 82, but getting to absolute wins produced would be trickier).  So WP predicts 48 wins, ezPM 45, RAPM 39, and APM 41.  And that ignores whatever Rubio would produce in 2750 minutes/5310 possessions.  An average player getting his minutes would produce about 5.7 wins, which would put the Wolves at anywhere from 44 to 54 wins, but we’ll leave Rubio out for now.

Any of these totals would be quite a jump for the Wolves, if not necessarily quite the 50 wins that Bill’s friend predicted.  Where does it come from?  WP says that the four new players would generate 17 wins and dumping the other four (Beasley, etc) would save them 4 wins for 21 of the 31 win change.  The other ten come from shifting minutes for the holdover players; primarily more for Love, less for Darko, Randolph, and Telfair.  ezPM, on the other hand, thinks that the four new players are only worth 22 points over the course of the whole season, which is two-thirds of a win.  Shipping out the other four saves nearly 400 points, or 12 wins.  So it places much more importance on the players leaving and shifting minutes (the other 15 wins gained) than on the players gained.  RAPM thinks the leaving players will save about 4.5 wins and the incoming players are essentially null.  In that case, 17 wins or so would come from rearranging minutes.  Finally, APM agrees with ezPM in that the leaving players are worth about 12 wins back to the team in cutting out negative production, and thinks the new guys are actually going to cost the team 2 wins.  So 14 wins are going to come from rearranging minutes.

Summing up: discounting whatever Ricky Rubio may contribute in a presumably high amount of playing time, the Bill Simmons trades would have improved the Wolves by anywhere from 22 to 31 wins depending on your metric of choice if all the players produce as they did this past season.  That’s kind of a wide margin, but when you’re moving up from 17 I don’t think the head office would complain.  The metrics do, however, also disagree on why the team would improve.  WP places a lot of the action on the incoming players whereas the other metrics think they are roughly a wash.  All the metrics agree that the leaving players will actually help the team (hooray for addition via subtraction!), but aren’t sure if it’s to the tune of 4 wins or 12.  And that means that simply reassigning playing time could account for anywhere between 10 and 17 wins in improvement just using existing players on the team.

Given that the trade never actually happened, this is a bad sign for the Wolves coaching staff.  What numbers should they believe, and who should get how many minutes?  Love and Tolliver are likely good contributors, but they play the same position.  Sadly, so is Anthony Randolph, who is a bit below average.  Ridnour is about average, so he gets a nod.  On the plus side, they did ship out Johnny Flynn and brought in Brad Miller, who is also about average.  Everyone else ranges from bad to terrible and appears to actually suck wins out of the team like basketball vampires.  The Wolves had better hope that Rubio was worth the wait and they get some help from Derrick Williams and Malcolm Lee (if I followed the trades correctly).  If you believe Arturo, Williams might be at least average.

So if the Wolves roll out a starting five of Miller, Love, Williams, Ridnour, and Rubio, they would have three average players, one very good player, and whatever Rubio can give them.  Even then, Miller only played 1000 minutes last year, so they might have to move Love to center and bring Tolliver in to play power forward fairly often.  Those are probably their two best line-ups, and they’d better hope they don’t need a third.  Every player they put in past those six start subtracting wins (and it’s possible Rubio will do that already).  It’s hard to imagine the Wolves being as bad again as they were last year, though, so things appear to be moving in the right direction even if GM Simmons didn’t get to make his moves.  As a wild guess, I’ll put the Wolves down for an optimistic 30 wins right now, and I hope I never have to write about them again.

About these ads
This entry was posted in Uncategorized and tagged , . Bookmark the permalink.

7 Responses to 50 Wins in Minnesota?

  1. EvanZ says:

    Alex, I think you have some decimals in the wrong place in that last table. I don’t predict hundreds of wins for Minnesota.

  2. EvanZ says:

    One thing I would say is that if I am predicting next season right now, I would assign a rating of -1.0 to all rookies. If Rubio and Willams get significant minutes, that team is not going to win 40 games.

    • Alex says:

      Yeah, I know rookies generally get a lower prediction than average. But I wanted to be optimistic in the hopes of giving the Wolves something to look forward to. It’s been kind of a depressing franchise, wasting most of Garnett’s career and then not getting anything to show for him. If Rubio and Williams are average even in two years, maybe the Wolves would be an average team with a couple other moves and a better rotation by then.

  3. wiLQ says:

    Will you check his other trades in the same way?
    Especially those with Suns, Iguodala and Bargnani?

    • Alex says:

      It took some work to get those things figured out, especially with the minute allotment, so I probably won’t do them in detail. But I can tell you that Dudley and Nash were more productive last season than anyone in trade 4 according to WP, ezPM, and APM (not using RAPM for 2011). Gortat is better than Pekovic (trade 5, and in the post). Bargnani is terrible by any metric (trade 9; although RAPM had him at about average rookie level in 2010). Iguodala does very well in WP and ezPM; APM has been high and low on him and RAPM had him as below average in 2010. In terms of his comment about Monta Ellis being better at basketball, that is not true for any metric I have in the past three seasons.

  4. Pingback: A Wolf’s Tale «

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s