Early Returns: Which teams drafted the best in creation?
Jan 29, 2021 11:23:27 GMT -5
delap, andrewluck, and 2 more like this
Post by jhb on Jan 29, 2021 11:23:27 GMT -5
I wanted to devise a method to fairly rank how each GM drafted in creation and then subsequently rank everyone's draft performance. Obviously, we're still in the early stages of the season and the sample size is small but I think we have enough data at this point that we can go ahead and start making some comparisons. Obviously, this doesn't account for the value of trades, just a ranking of the performance of each GM based on the total excess value created by all of the picks they made in slots where they ultimately drafted someone (does not account for post-creation trades) and then also ranking them by average excess value to account for differences in number of picks made.
First, I charted all 290 picks (including forfeits) and their WS/48 values (for players with at least 48 minutes, all other reported at 0) so far through a quarter season. Results below. I took a logarithmic trendline (equation below) that gave me the expected value in terms of WS/48 for each pick with a R2 value of .5188 so a semi-decent fit to the data considering the obvious talent disparity among GMs in this league.
From there I pivoted the results by team to see each club, the number of picks they had, their total excess wins created (let's call it WAR), the average excess wins created (let's call it AWAR), and then max and min values showing their best and worst pick in terms of wins created or lost.
To this point, we can see that the Mavericks obviously made the most of their many picks with the Bullets close behind. The Raptors, Grizzlies, Suns, and Hawks followed behind as the only teams to exceed .246 WS/48 over slot value in total. As a point of reference, Kareem Abdul-Jabbar, the greatest player in TMBSL 5.0 history by career WS/48, finished with a .298 WS/48 total for his entire career.
On a per selection basis, the Bullets come away so far as the winner of the draft. Funnily enough, compared to the other teams at the top this wasn't from having a couple of picks that jump off the page as great value selections, it mostly came from limiting downside and not making any truly bad picks. His worst pick was .038 WS/48 below expected value. Also want to give a shoutout to DaChap here as well, because while he didn't make any particularly great picks, he also made no really bad ones. Good work, rookie.
The best pick of the draft to this point belongs to the Mavericks with their selection of Moala Tautuaa. The 22nd pick of the draft would have been expected at this point to produce about .277 WS/48 but so far Moala is at .521. Kudos to Soup and the Mavericks' scouting department for this selection. The worst pick of the draft belongs to Majic and the Trailblazers. At 48th overall, we would expect Irwin Dambrot to be producing about .211 WS/48 so far, but he is at .015. Oof magoof.
Top 25 - Best Value Picks of Creation
Bottom 25 - Worst Value Picks of Creation
First, I charted all 290 picks (including forfeits) and their WS/48 values (for players with at least 48 minutes, all other reported at 0) so far through a quarter season. Results below. I took a logarithmic trendline (equation below) that gave me the expected value in terms of WS/48 for each pick with a R2 value of .5188 so a semi-decent fit to the data considering the obvious talent disparity among GMs in this league.
From there I pivoted the results by team to see each club, the number of picks they had, their total excess wins created (let's call it WAR), the average excess wins created (let's call it AWAR), and then max and min values showing their best and worst pick in terms of wins created or lost.
To this point, we can see that the Mavericks obviously made the most of their many picks with the Bullets close behind. The Raptors, Grizzlies, Suns, and Hawks followed behind as the only teams to exceed .246 WS/48 over slot value in total. As a point of reference, Kareem Abdul-Jabbar, the greatest player in TMBSL 5.0 history by career WS/48, finished with a .298 WS/48 total for his entire career.
On a per selection basis, the Bullets come away so far as the winner of the draft. Funnily enough, compared to the other teams at the top this wasn't from having a couple of picks that jump off the page as great value selections, it mostly came from limiting downside and not making any truly bad picks. His worst pick was .038 WS/48 below expected value. Also want to give a shoutout to DaChap here as well, because while he didn't make any particularly great picks, he also made no really bad ones. Good work, rookie.
The best pick of the draft to this point belongs to the Mavericks with their selection of Moala Tautuaa. The 22nd pick of the draft would have been expected at this point to produce about .277 WS/48 but so far Moala is at .521. Kudos to Soup and the Mavericks' scouting department for this selection. The worst pick of the draft belongs to Majic and the Trailblazers. At 48th overall, we would expect Irwin Dambrot to be producing about .211 WS/48 so far, but he is at .015. Oof magoof.
Top 25 - Best Value Picks of Creation
Bottom 25 - Worst Value Picks of Creation
