The Argument Against Signing Millsap/Jefferson/A Big Free Agent

So in my last post I was trying to find out how many wins we would have to find from a player in the 2014 free agency landscape to make signing Millsap a bad decision. The answer that I found was 4.6 wins, which translates to a thoroughly above average player. In this post I’ll be exploring the practicality of what I explored in the last post: if such a player is actually available in free agency, what the draft landscape actually is like (in other words, in my last post I just assumed that this draft would be average), how bad other teams will be next year in light of the draft and recent trades and how that affects the bobcats, and the effects of signing Millsap beyond a four year period.

A few notes. This post first started before CB’s insightful post about our position in free agency, but reading back on it, it kind of responds to a few of the points he made there, so just take that into account. Also, I realize that with the drafting of Zeller, it makes a lot less sense to sign Millsap, so this is really an argument against signing Al Jefferson or whatever FA is hot next (with the exception of Dwight, CP3 and MAYBE Bynum). The beauty of Al Jefferson is that his win shares are separated from Millsap’s win shares by .1 win, so really the calculations that we made with Millsap can be completely applied to Jefferson.

Let’s first address the issue of the bobcat’s record next year. There has been a lot of talk that the bobcats aren’t going to be one of the worst teams in the NBA next year (that they will be a mid lotto team rather than an early lotto team, so what’s the point in “tanking)…I’m here to dispel that notion. Unless we get very lucky, it is likely that we will be one of the three worst teams in the NBA next year. Now, I realize that this could seem like a very obvious statement for some, and an exaggerated statement for others, so for those people, we’ll look at the situation statistically in order to prove my point.

I see three ways that the bobcat’s record will increase (discounting luck): 1. the bobcat’s young players get better, 2. Zeller is really good, 3. the impact of a new coach. Lets first look at point #1. I personally believe that it is unrealistic to think that the bobcat’s players will progress at a pace far above other team’s young players (in other words a 21 year old on the bobcats should progress about as quickly as a 21 year old on the magic for example). So, I created a model to predict the average player’s improvements in win shares from year to year (so from 21 to 22 and from 22 to 23 and so on), and then I applied it to each individual player on each team (so young player’s win shares would increase and older player’s win shares would decrease). Then I added up the “new” calculated win shares for each team (so I added up the “new” win shares on the hawks for example). After that, I added up the old win shares from each team (so the win shares data from last year, unaffected by improvement from young players) and calculated the percentage difference between the old team win shares and the “new” team win shares. I used the Pythagorean method of calculating wins from points allowed and points scored (one of the most effective ways of predicting future wins, more effective than last year’s record) which is pretty much a de-lucker. Sometimes during a NBA season a team will outperform how well they are playing in the standings because of luck (that was charlotte this year by the way), the Pythagorean method ensures that this luck does not cloud our judgment when assessing how well teams actually played last year. Then I added one to the percentage difference I found earlier and multiplied each teams' Pythagorean wins by their percentage difference.

ESSENTIALLY FOR THOSE WHO GOT BORED WITH ME EXPLAINING ALL THAT MATH, I CREATED A THEORETICAL, LUCK NEUTRAL, NBA UNIVERSE WHERE THIS SEASON WAS PLAYED NEXT YEAR WITH THE EXACT SAME TEAMS AND PLAYERS WHO PROGRESSED ONE SEASON. ESSENTIALLY UNLUCKY TEAM’S RECORD SHOULD IMPROVE AND YOUNG TEAM’S RECORD SHOULD IMPROVE. Sadly when I did this, the bobcats had a grand total of 18.2 expected wins, worse that the second worst team, Orlando, by 5 wins. Now, this may seem a bit counter intuitive cause you’re probably thinking that because the bobcats are young, they should improve on their record from last year. The thing about this thinking is that it doesn’t recognize how LUCKY the bobcats got last year and how much of a major impact it had on our record. Sure this system I have created does have one of the largest percentage differences for the bobcats (a good thing indicating lots of young growth), the bobcats are such a bad, lucky team that it does not make that much of a difference. What you should get from this is that, even though the bobcats young players will get better, their improvement should be overshadowed by the return to luck normality for our record.

So, were left with two ways for the bobcats to catch up to the magic…the addition of Zeller and the addition of Clifford. Unfortunately, just from glancing at the stats, Zeller and Clifford shouldn’t be able to make up for 5 wins. As I explained in my last post, higher draft picks should result in better players, and that has been modeled by The average fourth pick should produce an average of roughly 5 wins a season in their first four years in the league. Ignoring the fact that Zeller should produce less than 5 wins his rookie season (because his production should theoretically go up, so for 5 to be the first number has to be below five) to simplify the calculations (it really shouldn’t have a great effect cause we are going to be doing this for all the teams in the draft), the players picked 1-3 should produce more than Zeller should. Let’s also ignore that fact as well and assume that all the players in the top four perform just as well as Zeller, to give an advantage to the bobcats in this calculation. The magic should also add a 5 win player, as should the cavs and the wizards. So unless Zeller is outperforming the top 3 picks (unlikely, as the model predicts that the value of picks goes down through the draft), we are not going to gain any ground on the teams closest to us in our predicted wins model. So the gap between us and the magic should still stay at 5 wins, us and Cleveland at 11 wins and us and Washington at 16 wins. The point is that even though we had a pick, all the other bad teams had picks close to ours, and it is unreasonable to expect our pick to be significantly better than theirs. By significantly better, I mean that Zeller would have to be as good as Millsap or Jefferson (around 7.5-8) wins to actually begin to make a difference, and even then, we are still predicted to be worse than the magic.

So…even with the addition of a draft pick, and the progression of all our young players, we’re still 5 wins down on the magic. To be even with the magic, the addition of Clifford would have to have a 25% positive effect on our record (which seems completely unreasonable). It is hard to actually find the real value of a head coach, but luckily we have some current events that may key us in on the true value of a great coach. Doc Rivers was just traded to the clippers for a 2015 first round draft pick (just keep in mind that Rivers’ trading value was completely inflated by the situation with Chris Paul). Anyway, I can’t imagine the clippers falling out of the playoffs in 2015 (and I don’t think the Celtics were counting on this when they made the trade), so I anticipate the pick being is the 16th pick, which has an average expected wins of 2.12. So, if a lot of things go right for the Celtics, Doc Rivers returned a value of 2.12 wins (which may have been inflated in the first place). I realize that Dunlap was a REALLY bad coach, but I think it is safe to say that he was as bad as doc is good, and unless we have really unearthed a gem with Clifford (and generally rookie coaches are not that successful and really need a few years to become above average coacher), it’s safe to say that he will be no better than an average coach next year. Really what I’m trying to say is that the addition of Clifford should be worth around 2.12 wins (if everything goes right).

Anyway…that brings us within 3 wins of the magic. If Zeller turns out to be Paul Millsap his rookie year, then we’ll have an expected record that is equal to the magic’s expect record…so we would have to get lucky again to not be the worst team in basketball. What I’m saying is that unless a ton of unlikely things happen then our current offseason additions/natural player progression, we should still be in the dredges of the NBA by season’s end (or at least that’s what the stats say).

Now that we have covered the bobcat’s potential improvement, we should look at the teams that could be getting worse next season. I’d say those teams are those who are highly connected to the free agent market: Philly, Bucks, Dallas, Boston, Utah, Lakers, Atlanta, and Houston. Now, of those teams, Philly (though it’s close), Bucks (assuming they match Jennings), Dallas, Boston, Utah, LA (assuming they don’t amnesty Kobe), Houston and Atlanta (assuming they match Teague) (so all of them) have better expected wins that us WITHOUT TAKING INTO ACOUNT FREE AGENTS. In other words, if the jazz were to roll out their team without al Jefferson, Paul Millsap, the rest of their free agents AND not sign anyone else, they would be statistically expected to be better than us (we are that bad). The hawks, with three players, have a higher expected wins than us. What I’m saying is that even if a team strikes out in free agency this year (unlikely because of the depth of this class), they will still be better than us, unless we’re lucky.

Anyway, so I know I said I would talk about more, but I’ll get to those things in a new fan post. The point of this is to show that the bobcats are really bad, and any talk of them being a respectable team without the addition of a free agent is completely overblown. Further the addition of a free agent would not be enough to completely kick them out of the lotto, but it could send them tumbling down in the order. Essentially the calculations done in the last post were mostly right (the number of wins needed to not sign al Jefferson/Millsap should be lower because our pick is predicted by this model to be lower than three, BUT, if one were to theoretically run the lotto millions of times the team with the #1 overall pick would get on average the 2.642th pick, which we can round to three). So the main takeaway is that the teams that are getting worse this year (as well as the teams that are staying bad) will not be bad enough to challenge our special brand of awful basketball.

Sorry for this seeming rushed, I was really tired when I did this/had a deadline because of internet connection so just ask if anything is really confusing. Also if anyone wants my spreadsheets that I did this on so you can actually see what I’m saying, instead of reading me ramble about my math, just ask for it in the comment’s section and I’ll try to figure out how to imbed in next week when I have internet again. I also have less statsy analysis of our situation if y'all want it, but it was the first draft of this post and kinda crummy.

Log In Sign Up

Log In Sign Up

Forgot password?

We'll email you a reset link.

If you signed up using a 3rd party account like Facebook or Twitter, please login with it instead.

Forgot password?

Try another email?

Almost done,

By becoming a registered user, you are also agreeing to our Terms and confirming that you have read our Privacy Policy.

Join At The Hive

You must be a member of At The Hive to participate.

We have our own Community Guidelines at At The Hive. You should read them.

Join At The Hive

You must be a member of At The Hive to participate.

We have our own Community Guidelines at At The Hive. You should read them.




Choose an available username to complete sign up.

In order to provide our users with a better overall experience, we ask for more information from Facebook when using it to login so that we can learn more about our audience and provide you with the best possible experience. We do not store specific user data and the sharing of it is not required to login with Facebook.