×
Message from Dave..... Moderator Approval

Don't panic if your post doesn't appear immediately.

× Rock Chalk Talk: Basketball

Anything pertaining to basketball: college, pro, HS, recruiting, TV coverage

Self is by far the winningest coach since 2004, real wins that is.

  • CorpusJayhawk
  • CorpusJayhawk's Avatar Topic Author
  • Offline
  • Platinum Member
  • Platinum Member
More
2 years 10 months ago #28170 by CorpusJayhawk
. This week I woke up one morning and you know how the secrets of the universe seem clearer when you first wake up in the morning? I get a lot of my ideas and inspiration in that netherworld of almost awake in the morning. Well one thing that has bothered me since I showed it last year was the total wins by college program from 2004 to present. Gonzaga and Mark Few just passed Bill Self last season. That really sticks in my craw because Self has played a robustly more challenging and difficult SOS than Gonzaga. So Gonzaga winning more games but playing a radically easier schedule just does not seem fair. Well being an engineer back in the day, like all engineers we love to normalize. We can normalize any data set against any variable. It is genetic. So I wanted to transcend a simple sum of victories when comparing various programs across some period of time. And I did in a most satisfying and brilliant manner. I do a lot of this rabbit trail kind of stuff, almost always for my own amusement. But I thought you might enjoy this one since, like me, you are all KU Jayhawk and Bill Self fans. So first, let me show you the data that was bothering me. This is a list of wins going back to and beginning with the 2004 season when Self arrived at Kansas. Now if you are going to check me on my numbers you will see that my win totals don't match the official win totals. That is because I am only using Division 1 games. I have culled out the non-D1 games from this calculation



This is the top 50 winningest teams since 2004. You can see at the top all the usual suspects. There are no surprises. But No. 1 with 538 wins (through January 18th) is Gonzaga. Okay, fine. Gonzaga has won a lot of games. They have won 1 more game than Kansas and Bill Self. But good grief! Look at their SOS. Now let me explain briefly what the SOS rating is exactly. My SOS rating is a meaningful number. It is the percent of games the average team in the country would lose if they played the exact same schedule as the team in question. So for instance, if the average team in the NCAA Div 1 played the exact same schedule as KU played from 2004 to present, it would have lost 61.2% of those games. That same average team would have lost only 49.2% of the games on the Gonzaga schedule. So the SOS is a pretty intuitive and meaningful number. Incidentally, and this is important, the 0.612 SOS of Kansas is the highest SOS in Div 1. In other words, KU has played the toughest SOS since Bill Self has been at Kansas. So here is what bugs me. Many people might look at this win record and say "gosh, Gonzaga has won the most games. That means they have been the best team." But Gonzaga has won more games because they have played a much easier schedule. So how can we compare KU and Gonzaga and everyone else on an SOS adjusted manner? Well I have exactly that solution an you are lucky because only here at the RCC will you get such incredible comprehensive analysis. The answer Net Wins. What is Net Wins you ask? It is pretty straightforward. You calculate the number of wins the average team NCAA Div 1 would have if they played the same schedule as each team. Then you subtract that from the actual number of wins that team had. So for instance, in the 658 games KU has played since 2004, if the average team in Div 1 would have played that exact same schedule they would have won only 255.1 games. Yet KU won 537 games. Thus, KU has won 281.9 more games than the average Div. 1 team would have won. This is the "Net Wins". It uses the exact same standard for all teams so we can now compare the records for al teams adjusting exactly for the SOS. As you have hopefully guessed by now, Kansas leads the pack by a mile. Duke is a distant second and UNC is an even more distant 3rd. There is no doubt that when it comes to winning games, since 2004, Bill Self and Kansas are the gold standard. Gonzaga is a very respectable 5th place 67 games behind Kansas. The fact that Gonzaga, Memphis VCU, Wichita St. Saint Mary's, Belmont , and BYU all appear in the top 25 in total wins since 2004 is very misleading since they play in conferences where many of their games are against much lower rated teams. Not only that, even many Power 6 conference teams do not play the strength of schedule that KU plays year in and year out. So it is nice that now we have a way to adjust this win total to the actual SOS that teams play. Once you make this adjustment, only Gonzaga and Memphis remain in the top 25 teams as non Power 6 conference teams. And honestly, KU's net win margin is simply crazy. We should drink this data in and realize what a blessing it has been to have KU and Bill Self leading this team. Here are the top 50 in Net Wins.



I wanted to share that with you. It is a real thing. It is a solid fundamental analysis that is using real data and is a much better measure than simply using total wins. And it shows that KU and Bill Self have dominated college basketball since 2004. I mean seriously, between Kansas and Duke, the domination is just ridiculous. UNC, Kentucky, Gonzaga, Villanova and Michigan St. have all done so much better than the pack, but KU and Duke, and especially KU has so dominated winning that is is amazing. I hope you fine that as fun and satisfying as I do. And just in case you are wondering who else had played a tough schedule, here is a ranking of the toughest SOS since 2004. And just in case you were wondering, Gonzaga has played the 87th toughest SOS in this time period. So our beloved Jayhawks have played the toughest schedule and also have won the most games adjusted for SOS. Nice!!


Don't worry about the mules, just load the wagon!!
The following user(s) said Thank You: HawkErrant, hairyhawk, Socalhawk, porthawk

Please Log in or Create an account to join the conversation.

Share this page:

 

More
2 years 10 months ago #28172 by hairyhawk
I think that is a very good analysis and as all of us know in a small sample size statistics can be skewed but in the long run they are very accurate. all Di games since 2004 is a large sample size. That is both impressive coaching and impressive analysis. On another note I am still anxious to see your player evaluation for last game as even though Mr. Agbaji only scored 10 points I thought he really helped the team but I could be way off.

Please Log in or Create an account to join the conversation.

Powered by Kunena Forum