{"id":461,"date":"2011-04-13T10:21:08","date_gmt":"2011-04-13T15:21:08","guid":{"rendered":"http:\/\/godismyjudgeok.com\/DStats\/?p=461"},"modified":"2011-04-13T10:23:51","modified_gmt":"2011-04-13T15:23:51","slug":"with-or-without-you-okc-perk-and-nazr","status":"publish","type":"post","link":"http:\/\/godismyjudgeok.com\/DStats\/2011\/nba-adjusted-efficiencies\/with-or-without-you-okc-perk-and-nazr\/","title":{"rendered":"With or Without You: OKC, Perk, and Nazr"},"content":{"rendered":"

In continuing my series of With-or-Without-You (WOWY) analyses, I will next look at my hometown Oklahoma City Thunder, one of the hottest teams going into the playoffs.\u00a0 In the first two posts on WOWY, I looked at Oklahoma City in January<\/a> (before the trades), and yesterday I looked at the Bulls and their strength<\/a> going into the playoffs.\u00a0 In the Bulls post, I also revised and expanded this method with regression to the mean and Bayesian weighting for best future prediction.<\/p>\n

Since January, the Thunder traded away Jeff Green<\/a> and Nenad Krstic<\/a>, both of whom looked very bad from a WOWY perspective, and added Kendrick Perkins<\/a> and Nazr Mohammed<\/a> to the regular rotation.\u00a0 After the trades, the Thunder have gone on a tear, going 19-4 since the trades came through and 13-3 since Kendrick Perkins<\/a> entered the lineup.\u00a0 Of course, the opponents have been rather weak over the last 23 games, with the average game being about 1 point easier than average.\u00a0 There have been 3 huge wins since the trade, however: at Miami, at Denver, and at the Lakers–those were the 3 hardest games since the trades and the Thunder won them all.<\/p>\n

Let’s start out this analysis with the normal charts:<\/p>\n

\"\"<\/a>

Oklahoma City Thunder Performance Chart<\/p><\/div>\n

\"\"<\/a>

Oklahoma City Thunder Efficiencies Table<\/p><\/div>\n

From the team performance chart, we can see the strong upward trend the Thunder have been showing.\u00a0 In the efficiencies table, we can see who was missing for each game, how difficult the game was, and how the team did on offense, defense, and overall.\u00a0 Like I mentioned in the last OKC WOWY post<\/a>: Nick Collison<\/a> is important to this team!<\/p>\n

Okay, let’s get down to the nuts-and-bolts of WOWY: first, the raw regression, no weighting, assigning value to each player that’s missed at least 3 games and is a regular otherwise.<\/p>\n\n\n\n\n\n\n\n\n\n\n\n
<\/th>\nEff Mar<\/th>\nOff Eff<\/th>\nDef Eff<\/th>\nG In<\/th>\nG Out<\/th>\nMPG<\/th>\n<\/tr>\n
Team<\/strong><\/td>\n6.2<\/strong><\/td>\n5.9<\/strong><\/td>\n-0.3<\/strong><\/td>\n<\/td>\n<\/td>\n<\/td>\n<\/tr>\n
Kevin Durant<\/a><\/td>\n5.8<\/td>\n11.5<\/td>\n5.7<\/td>\n77<\/td>\n4<\/td>\n39.1<\/td>\n<\/tr>\n
Jeff Green<\/a><\/td>\n-4.1<\/td>\n-4.1<\/td>\n0.0<\/td>\n49<\/td>\n32<\/td>\n37.0<\/td>\n<\/tr>\n
Thabo Sefolosha<\/a><\/td>\n4.6<\/td>\n1.1<\/td>\n-3.5<\/td>\n78<\/td>\n3<\/td>\n26.0<\/td>\n<\/tr>\n
Kendrick Perkins<\/a><\/td>\n2.1<\/td>\n0.9<\/td>\n-1.2<\/td>\n16<\/td>\n65<\/td>\n25.5<\/td>\n<\/tr>\n
Nenad Krstic<\/a><\/td>\n0.7<\/td>\n3.3<\/td>\n2.6<\/td>\n47<\/td>\n34<\/td>\n21.7<\/td>\n<\/tr>\n
Nick Collison<\/a><\/td>\n8.2<\/td>\n5.6<\/td>\n-2.6<\/td>\n71<\/td>\n10<\/td>\n21.5<\/td>\n<\/tr>\n
Nazr Mohammed<\/a><\/td>\n0.8<\/td>\n0.3<\/td>\n-0.5<\/td>\n23<\/td>\n58<\/td>\n17.3<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

(Note the nicer formatting!\u00a0 Thanks, ASAP utilities<\/a>!)<\/p>\n

What do we see? First of all, the lack of stabilization in the regression shows: KD can’t hurt the team by 6 points per 100 possessions on defense!\u00a0 That’s the small sample size coming through.\u00a0 For the players with a larger sample size, the results look pretty reasonable, though Nick Collison<\/a> again looks like a superman.<\/p>\n

Let’s look at the stabilized results, regressing towards average within the regression:<\/p>\n\n\n\n\n\n\n\n\n\n\n\n
<\/th>\nEff Mar<\/th>\nOff Eff<\/th>\nDef Eff<\/th>\nG In<\/th>\nG Out<\/th>\nMPG<\/th>\n<\/tr>\n
Team<\/strong><\/td>\n8.7<\/strong><\/td>\n7.4<\/strong><\/td>\n-1.3<\/strong><\/td>\n<\/td>\n<\/td>\n<\/td>\n<\/tr>\n
Kevin Durant<\/a><\/td>\n2.9<\/td>\n2.8<\/td>\n-0.1<\/td>\n77<\/td>\n4<\/td>\n39.1<\/td>\n<\/tr>\n
Jeff Green<\/a><\/td>\n0.7<\/td>\n0.7<\/td>\n0.0<\/td>\n49<\/td>\n32<\/td>\n37.0<\/td>\n<\/tr>\n
Thabo Sefolosha<\/a><\/td>\n2.2<\/td>\n1.3<\/td>\n-0.9<\/td>\n78<\/td>\n3<\/td>\n26.0<\/td>\n<\/tr>\n
Kendrick Perkins<\/a><\/td>\n2.6<\/td>\n1.7<\/td>\n-1.0<\/td>\n16<\/td>\n65<\/td>\n25.5<\/td>\n<\/tr>\n
Nenad Krstic<\/a><\/td>\n0.2<\/td>\n0.8<\/td>\n0.6<\/td>\n47<\/td>\n34<\/td>\n21.7<\/td>\n<\/tr>\n
Nick Collison<\/a><\/td>\n2.9<\/td>\n1.7<\/td>\n-1.1<\/td>\n71<\/td>\n10<\/td>\n21.5<\/td>\n<\/tr>\n
Nazr Mohammed<\/a><\/td>\n2.6<\/td>\n1.6<\/td>\n-1.0<\/td>\n23<\/td>\n58<\/td>\n17.3<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

The stabilization really shows, here.\u00a0 Nick Collison<\/a> is still really good per minute (that +2.9 translates approximately to a +\/- of 2.9 as well), while both Jeff Green<\/a> and Nenad Krstic<\/a> at least make it above replacement level!\u00a0 Both of them have a translated +\/- of below -2.5, though.\u00a0 KD looks rather average, probably because the small sample size has trouble overcoming the regression to the mean, but one thing is clear: he’s better on O than on D.<\/p>\n

Finally, let’s look at the Bayesian results, with more recent results weighted more heavily than older ones.\u00a0 This is more useful for predicting the team future performance than quantifying the impact of individual players.\u00a0 Here’s that table:<\/p>\n\n\n\n\n\n\n\n\n\n\n\n\n
<\/th>\nEff Mar<\/th>\nOff Eff<\/th>\nDef Eff<\/th>\nG In<\/th>\nG Out<\/th>\nMPG<\/th>\n<\/tr>\n
Team<\/strong><\/td>\n9.9<\/strong><\/td>\n8.3<\/strong><\/td>\n-1.5<\/strong><\/td>\n<\/td>\n<\/td>\n<\/td>\n<\/tr>\n
Kevin Durant<\/a><\/td>\n3.0<\/td>\n2.5<\/td>\n-0.5<\/td>\n77<\/td>\n4<\/td>\n39.1<\/td>\n<\/tr>\n
Jeff Green<\/a><\/span><\/td>\n1.2<\/td>\n1.1<\/td>\n-0.1<\/td>\n49<\/td>\n32<\/td>\n37.0<\/td>\n<\/tr>\n
Thabo Sefolosha<\/a><\/td>\n2.2<\/td>\n1.3<\/td>\n-0.9<\/td>\n78<\/td>\n3<\/td>\n26.0<\/td>\n<\/tr>\n
Kendrick Perkins<\/a><\/td>\n2.8<\/td>\n1.9<\/td>\n-0.9<\/td>\n16<\/td>\n65<\/td>\n25.5<\/td>\n<\/tr>\n
Nenad Krstic<\/a><\/span><\/td>\n0.4<\/td>\n0.8<\/td>\n0.4<\/td>\n47<\/td>\n34<\/td>\n21.7<\/td>\n<\/tr>\n
Nick Collison<\/a><\/td>\n2.7<\/td>\n2.0<\/td>\n-0.7<\/td>\n71<\/td>\n10<\/td>\n21.5<\/td>\n<\/tr>\n
Nazr Mohammed<\/a><\/td>\n2.9<\/td>\n1.8<\/td>\n-1.0<\/td>\n23<\/td>\n58<\/td>\n17.3<\/td>\n<\/tr>\n
Team Future Pred.
\n<\/strong><\/td>\n		8.2<\/strong><\/td>\n6.3<\/strong><\/td>\n-1.9<\/strong><\/td>\n<\/td>\n<\/td>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

So here we go.\u00a0 The prediction for the team, if everybody is healthy and Jeff Green<\/a> and Nenad Krstic<\/a> stay gone, is for a +8.2 efficiency differential<\/strong>.\u00a0 That is scary good.\u00a0 Not as scary good as the Bulls’ +10.0 going forward, but that +8.2 is better than any team has maintained for the whole year this year (Chicago leads with a +7.38 average).<\/p>\n

In other words, the Oklahoma City Thunder must be considered a contender<\/strong> since the trades.\u00a0 They ditched their weakest links and added in 2 good contributors.\u00a0 (Wrapped into the numbers is the value of starting Serge Ibaka<\/a>–his quality is what makes Jeff Green<\/a>‘s numbers look so bad!)<\/p>\n

I’ll try to take a look at some of the other contenders with this approach as well.\u00a0 One cautionary note: OKC, like Chicago, will have less of a beneficial effect from shortened playoff rotations than some of the older and thinner teams–that’s a hard effect to quantify, but it favors the Lakers, Celtics, and Heat.<\/p>\n","protected":false},"excerpt":{"rendered":"

In continuing my series of With-or-Without-You (WOWY) analyses, I will next look at my hometown Oklahoma City Thunder, one of the hottest teams going into the playoffs. In the first two posts on WOWY, I looked at Oklahoma City in January(before the trades), and yesterday I looked at the Bulls and their strength going into the playoffs. In the Bulls post, I also revised and expanded this method with regression to the mean and Bayesian weighting for best future prediction.<\/p>\n","protected":false},"author":3,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"ngg_post_thumbnail":0,"footnotes":""},"categories":[10,12,15],"tags":[21,54,37,31,22,24,58,25],"_links":{"self":[{"href":"http:\/\/godismyjudgeok.com\/DStats\/wp-json\/wp\/v2\/posts\/461"}],"collection":[{"href":"http:\/\/godismyjudgeok.com\/DStats\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/godismyjudgeok.com\/DStats\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/godismyjudgeok.com\/DStats\/wp-json\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"http:\/\/godismyjudgeok.com\/DStats\/wp-json\/wp\/v2\/comments?post=461"}],"version-history":[{"count":13,"href":"http:\/\/godismyjudgeok.com\/DStats\/wp-json\/wp\/v2\/posts\/461\/revisions"}],"predecessor-version":[{"id":476,"href":"http:\/\/godismyjudgeok.com\/DStats\/wp-json\/wp\/v2\/posts\/461\/revisions\/476"}],"wp:attachment":[{"href":"http:\/\/godismyjudgeok.com\/DStats\/wp-json\/wp\/v2\/media?parent=461"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/godismyjudgeok.com\/DStats\/wp-json\/wp\/v2\/categories?post=461"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/godismyjudgeok.com\/DStats\/wp-json\/wp\/v2\/tags?post=461"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}