FanPost

How many wins can this team produce?

Alright. It looks like we might be set, pending a surprising move. I'd hoped there might be another big move, but it might be waiting until the season or even the trade deadline now. So it looks like Alan Anderson, Aaron Gray, and John Lucas III are the final pieces to this roster. The question is, if this is the final form of the team, how will it perform?

I'll use WP/48 and WS/48 to build some guesses. Now, this is far from perfect, as WP and WS tend to fluctuate when a player changes teams (and even when they don't, if teammates, coach or role change), but it's the best we've got. Here are the values I have for the lineup as it stands (with some projecting for rookies).

23904419_medium

So. This will depend on the minutes breakdown. I've gone through and set up two different scenarios - a perfect season where there are no injuries and everyone plays the minutes they would in an ordinary rotation, and a realistic scenario where I allocated how many games a player might miss based on last year's injuries (the last few years for those players who missed a lot of time last year). I went through and calculated how much additional time this would give the reserves and included that in their MPG averages. Note that these are not the minutes I would give players, but the way I expect it to go down (or be planned to go down, anyway).

For the rookies, I used their Draftexpress best and worst case scenarios to determine their WS and WP rates. I used the lowest of the rookie WS/48's of their best and worst cases, and the lowest of their best and worst cases' WP/48 this past year, prorated to their rookie year based on the ratio of their WS now versus then, as I only know how to access this year's WP numbers. Simply put, Acy's WP and WS came out to 0, and so did Ross' WP.

Minutes_medium

So. How do the two scenarios work out? Here is the summary table - they are actually very close.

Resultspb_medium

Seems like a 8th-9th place finish is in our near future, as many have speculated. Now, an interesting exercise would be to try the various minute breakdowns that people might want to see. Just give me the minutes you want certain players to play, and I'll plug them in (I'll use the "Ideal" approach, as it is much faster and seems like a fair estimate of the "Actual" approach). Or if you are interested in questions like - what would be the best case scenario? What if Bargs played 82 games like he played 13? Let me know, I'll try to find the data and compile it.

************************************************************************************************************************

Okay, I'm trying another thing here. The above values are based on individual defense and offense - very good at catching what the player does, and what the player he is guarding is doing. However, they do ignore the overall impact on the team. Does the team score more points (relative to what they give up) when the player plays than when he sits? What about the strength of his teammates, and who the opponents are?

I'm turning to a couple of stats - adjusted plus minus, and regularized adjusted plus minus. Adjusted plus minus takes the player's raw plus minus, then factors in things like pace, which teammates he played most with, which opponents he played against most, etc. Regularized adjusted plus minus does the same, but with a sort of smoothing function, that avoids the jumping around that adjusted plus minus can have. More predictable, and therefore reliable, but less able to track immediate trends in a player's play.

Apm_medium

So, using those values, I established what the team's offensive and defensive ratings would be had the player played all 48 minutes for all 82 games with completely average players (0 impact on DRTG and ORTG). I did this for 1-year adjusted plus minus, 2-year adjusted plus-minus, and 3-year regularized adjusted plus minus. These can then be punched into the pythagorean win calculation:

Win % = ORTG^16.5 / (ORTG^16.5 + DRTG^16.5)

So, then I applied the player's win percentage to all 82 games, and found a Pythagorean WIn value. The table below lists the case by case pythagorean win value for each player. Note that rookies or players with missing data (like Anderson the past 2 years) are assumed to be "average" players with a PW/48 of 0.100 (or a 0 +/-). Keep in mind that the most recent year is best described by the "1 year adjusted" value, while the 3 year regularized value is probably the least precise, but the most accurate - doesn't tell you that much about recent performance, but is more reliable. The average gives a nice balance.

Pw48_medium

The final chart compiles the team's total PW, using the minute breakdown I listed above.

Pwtotals_medium

Nice little uptick in wins there, relative to the individual performance based predictions. Very much in the same range, so that lends even more credence to the idea that this team might just make the playoffs. Once again, if you have any suggestions for different minute breakdowns, I can easily calculate the totals.

X
Log In Sign Up

forgot?
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,

Join Raptors HQ

You must be a member of Raptors HQ to participate.

We have our own Community Guidelines at Raptors HQ. You should read them.

Join Raptors HQ

You must be a member of Raptors HQ to participate.

We have our own Community Guidelines at Raptors HQ. You should read them.

Spinner

Authenticating

Great!

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.

tracking_pixel_9347_tracker