[$Ball]Quantifying Next Year's Success(or Failure)

First off, I feel like I need a big fat disclaimer because most of what I am about to talk about is hypothetical, and I fully realize basketball is not played on spread sheets or what ifs. However, to be able to analyze anything at all means we must make some assumptions. That doesn't mean it has to be taken literally. And please, somebody correct me if any of my #s or, more importantly, logic is wrong.


So the essence of Moneyball is the optimal usage of resources. In this case, it is the attempt to squeeze every penny of out of a cap constrained payroll. Actually it is to get the most value out of those pennies. In other words, the whole point of $Ball is to acquire players that will outperform their salary. This way, when faced against a team with similar payroll, $Ball wins.

My question is, will we outperform(our payroll) next year? And how will we know?

This question has been irrelevant in previous years. This is because we've had a full roster, and full payroll. The measure for success has been simple: Win a Championship. In reality, we were still asking that same question, only it was unspoken- outperforming a full payroll means outperforming every other team = Win a Championship.

This year, however, half our payroll will be inactive. Obviously we can not measure success by a Championship. Not only is that unrealistic, it is illogical: we are not on the same playing field as other teams since we do not have the same active payroll. So how do we measure success? I've used two ways to quantify "outperformance", yielding surprisingly similar conclusions. This is all very hypothetical though so bare with me.

Here are some numbers first:

Houston Rockets 09/10 active payroll: $34,802,866
This is our total payroll minus the salaries for Yao Ming and Tracy Mcgrady. Also assuming we cut James White, while signing David Andersen for $2.5mil and Chase Budinger, Jermaine Taylor for $800,000 each.

NBA average team payroll: $63,565,527
This is just an average of every team's payroll. I realize it may be a little off since some players have yet to sign. However for arguments sake, we'll use it anyways.

http://hoopshype.com/salaries.htm


Before we proceed, we need to determine how to quantify performance. This is easy: wins.

So now, to determine "outperformance" we need to first determine "par". If you put NBA team wins on a Bell curve, par would be 41. Another way to look at it is if every team were equal, then they would all have 41-41 records. We have already found out what the par payroll is: the average salary of $63,565,527.

With those numbers we can extrapolate par for the Rockets.

The ratio of our active payroll to par payroll is:
$34,802,866/$63,565,527= 0.5475

Therefore the par number of wins for the Rockets is:
0.5475*41= 22.45

Outperformance next year means we must win more than par, or 22 games.

_________________________________________________________________


A second way I determined this was even more simple.

Total salary for all teams last year: $1,906,965,795

Total number of wins every year: 1260

From this we can calculate, in essence, on average how much $ it costs per win.

$1,906,965,795/1260= $1,513,464.91

Since we know the Rockets active payroll, we can now determine par wins:
$34,802,866/$1,513,464.91= 23.00

This is a very similar result compared to the first method. Outperformance here means winning more than 23 games.

Now that we know this, what does it all mean? It means I have way too much time on my hands, and if you've read this far and you are still interested, you do too.

Ok, seriously now. We've determined that par wins next year is 22-23 wins. What that means is if we have no edge what so ever, no advantage or disadvantage in any way, shape or form, we win 22-23 games. What it also means is that our players have all performed to their salaries.

However the purpose of $Ball is to find players that outperform their salaries. We can conclude that we must win more than 22-23 games next year, or $Ball has failed. That is our quantified measure of success, on a level playing field.

Now, I readily admit that there are plenty of holes in this argument, because there are other factors that affect performance. Coaching and chemistry immediately come to mind. So if we win significantly more than 22-23 games next year, these factors have to be taken in consideration. However, at least partly it will be because our players have outperformed their salaries due to the principles of $Ball.

 

Keep in mind I am using a very elementary model here, so don't take it too seriously. We neither have Morey's resources nor his time and effort. I'm sure he has black boxes figuring this stuff out somewhere. LOL.

 

I don't know if I would call that a flaw, it is an assumption. I cannot give you an active payroll for every team because injuries are unforeseen. To have any analysis or discussion at all we need some constants.

 

Actually, I apologize, I see what you are saying now. You are right, for accuracy's sake I should have used the average active payroll. It's just too tedious to go through every team figuring out who is currently injured though. If you would like to do it and provide me the #s, we can plug it back in the formula though.

Again, I wouldn't say it is a HUGE flaw. I doubt it would change the average payroll # I used by more than 2-3mil. I mean, there aren't that many other teams missing 40mil from their roster.

 

The easiest way is just to go through every single team, see who is currently injured, and sum up the active payroll. Then take the average.

I say easiest but I sure as hell ain't doing it. Someone else can.

Like I said though, the outcome won't be affected by that much, however it will be more accurate.

 

Oh, that's not what I am saying at all. The conclusion was that if our players play to their salaries, we will win 22-23 games. However I have a sneaking suspicion that they will outperform by a large margin. That is the point of $Ball after all.

 

This is a great point, and by far the biggest question mark to my whole premise. Here was my rationale:

On a case by case basis, there is absolutely NOT a linear relationship. All you need to look at is teams like the Knicks. They have a 100mil payroll yet can't win squat. Certain teams will get more out of their payrolls and certain teams, like the aforementioned, get very little.

However, I figured taken as a whole, the discrepancies would even out, and form a more linear relationship. When all things are equal, more salary = more talent = more wins. Of course that qualifier is paramount.

 

In a modified setting, absolutely.

 

And to further elaborate, even though we can't be sure if it is a linear relationship, we can easily assume it is a positive relationship. If that is the case, my conclusions can at least serve as guesstimations, as opposed to accurate predictions. Which were my intentions anyways.


BTW guys sorry for triple posting, I guess I am the only one up so far