Was Cobbins Really a Big Lose?

[SoonerBounce13]

He had played only 13 games this year. He fouled out in quick order in 2 of those and was hurt 3 minutes in in another. In short, a guy that shows steady improvement from 5 and 6 to 7 and 6 is not going to just regress back to 4 and 4. You are working with a small sample size and the fact he was averaging 4 and 4 does not change the player that he is.

I don't think you are quite understanding. His stats per minute played are very similar to his career numbers. I"m not saying he got worse. I'm saying he is the same player as last year and that same player is overrated. It wasn't losing him that made osu not a ff team, it was depth.

I notice you didn't respond to the point that losing Hield last year (an 8 and 4 guy) was a huge blow for OU.

Explain how it was a huge blow?He lost in OT on the road to OSU and Texas. We handly beat tech, baylor, and WVU. He came back and we lost our next three games to TCU and iowa state and SDSU. It wasn't a huge loss last year because we had depth and a guy who can coach.

I also notice that you have yet to account for the fact that there are other aspects to basketball than what shows up in the points and rebounds column.

I never said anything to the contrary. in fact I even said something about not being able to really measure defense.

His loss affected OSU a significant amount and to argue otherwise is just foolhardy.

you really think OSU would have been in a lot better position with cobbins than they are today? And again, it's not the loss of cobbins, but the loss of a body b/c they lack depth

 

[oukyle]

I don't think you are quite understanding. His stats per minute played are very similar to his career numbers. I"m not saying he got worse. I'm saying he is the same player as last year and that same player is overrated. It wasn't losing him that made osu not a ff team, it was depth.

Again, who outside of Stillwater claimed OSU was a sure fire Final 4 team in the preseason? A contender? Yes. But they were preseason 8 and 12. That's a Sweet 16 type team.

Explain how it was a huge blow?He lost in OT on the road to OSU and Texas. We handly beat tech, baylor, and WVU. He came back and we lost our next three games to TCU and iowa state and SDSU. It wasn't a huge loss last year because we had depth and a guy who can coach.

I don't know, I saw a lot of worry about where our season was headed here:
http://ouhoops.com/forum/showthread.php?t=27932

Also, we lost 2 games in OT without Hield (think he makes even a 1 point difference in those games? I do.) and a close loss against TCU after he returned but was ineffective. You could make a really good claim that we would have won all 3 of those games with Hield and certainly 2 of 3. With 2 or 3 extra wins instead of losses, we aren't a 10 seed. We would have had a much better seed and a much better matchup. So, yeah, Hield getting hurt was a pretty big loss.

I never said anything to the contrary. in fact I even said something about not being able to really measure defense.

I am glad to see that you understand that.

you really think OSU would have been in a lot better position with cobbins than they are today? And again, it's not the loss of cobbins, but the loss of a body b/c they lack depth

Would OSU be in a better position? Absolutely. Is he the biggest issue with their team? Hell no.

 

[earlpearl]

According to Kenpom, he was a pretty significant loss....at least with respect to their defense. OSU's adjusted Def rating is currently at 51. Their adjusted Def rating in late December/early January was top 20. Their eFG% (effective FG% defense) has also worsened over the same time period. Cobbins was their most consistent rim protector and by not having him in there, teams are more easily getting to the rim....at least that is what you can conclude from the stats.

Exactly. Cobbins was good for four blocks a game. That is a potential 8 point swing. Not to mention the the other 4 or 5 shots that he altered. Defensively he was the only interior player that had any kind of physicality. He could bang and play position defense. Cobbins made us a legitimate final four threat. I will say this however, the guard play has been horrendous since conference play. The loss of Cobbins alone can't explain the total implosion of this team. The guards have to play better. The final piece is that even talented players must have a coach that coheses them together with discipline. Ford is not that guy. A talented team without a coach might just bow out in the first round even with Cobbins.

 

[StatGeek]

Exactly. Cobbins was good for four blocks a game. That is a potential 8 point swing. Not to mention the the other 4 or 5 shots that he altered. Defensively he was the only interior player that had any kind of physicality. He could bang and play position defense. Cobbins made us a legitimate final four threat. I will say this however, the guard play has been horrendous since conference play. The loss of Cobbins alone can't explain the total implosion of this team. The guards have to play better. The final piece is that even talented players must have a coach that coheses them together with discipline. Ford is not that guy. A talented team without a coach might just bow out in the first round even with Cobbins.

Sighting Kenpom on this is next to meaningless. The overwhelming reason the defensive efficiency dropped was the competition, not Cobbins. Oh, and can we correct the title of this thread, it's hard to look at.

 

[stormspencer]

Sighting Kenpom on this is next to meaningless. The overwhelming reason the defensive efficiency dropped was the competition, not Cobbins. Oh, and can we correct the title of this thread, it's hard to look at.

Wrong wrong wrong. Kenpom has many determinants that go into his rankings. One of those being SOS (who you played). His metrics take into account not only points per possession (offense and defense), but in identifying associated factors and weighing those rankings based upon that. For example, he includes parameters/indicators like "how good of a team you're playing (competition)", pace of play, and a "luck" factor.....and all of those indicators go into determining your offensive and defensive efficiency.....and thus your ranking.

 

[StatGeek]

Wrong wrong wrong. Kenpom has many determinants that go into his rankings. One of those being SOS (who you played). His metrics take into account not only points per possession (offense and defense), but in identifying associated factors and weighing those rankings based upon that. For example, he includes parameters/indicators like "how good of a team you're playing (competition)", pace of play, and a "luck" factor.....and all of those indicators go into determining your offensive and defensive efficiency.....and thus your ranking.

Yes , but does it do so well? Do you understand the formula and it's flaws?

 

[stormspencer]

Yes , but does it do so well? Do you understand the formula and it's flaws?

I understand the formula and it's flaws to a degree. The main issue in his formula that he has tried to account for over the last several seasons is the "Wisconsin problem". Here is a link http://www.hagrin.com/346/kenpom-wisconsin-problem

The main issue that I've noticed just by looking over his top 30 is that teams with good records that play at a slower pace (Wisconsin, Pitt) tend to be overvalued in his rankings. The other two issues with Kenpom are:
1) Problems at the top are significantly more magnified than problems near the bottom.
2) The ability to win is somewhat undervalued (and this is an issue that affects OU this season because we have won a lot of close games)....and thus I think we are "undervalued" in his system/we should be ranked higher.

If you are a "stat-hound" your name suggests "Statgeek", I would recommend spending the $20 for a year's subscription. You get a really good explanation of all the formulas and how it works. I think it's awesome, but it is by no means a perfect system. Still, you will find that Vegas lines are usually pretty darn close to his rankings when you account for home/away +/- points in factoring favorites/underdogs.

 

[MNsooner]

But Vegas lines aren't meant to predict the outcome of the games. They are meant to balance the betting. If his lines are close to the vegas lines then that is more of an indictment of his system than praise.

 

[garyeb2]

PT Barnum had a famous quote that might be relevant.

 

[stormspencer]

But Vegas lines aren't meant to predict the outcome of the games. They are meant to balance the betting. If his lines are close to the vegas lines then that is more of an indictment of his system than praise.

That's a fair point MN, but unlike football, college basketball has a larger sample size (more games) to formulate more metrics from and incorporate into simulations/rankings. As the season progresses, Vegas will utilize these "multi-simulations" of upcoming games in order to assist in setting their lines. Additionally, you will find more opportunities (no matter how good the team) of teams having "flat spots"....and Vegas knows how to sniff that out and set their lines accordingly.

One last thought regarding basketball/football. You won't find heavy money coming in late/early on a college basketball powerhouse as often as you do on a college football powerhouse because of:
1) Lines aren't usually set until around 24 hours before the game.
2) The margin of error is greatly diminished in college bball because of the larger sample size of data.

 

[stormspencer]

PT Barnum had a famous quote that might be relevant.

"There's a sucker born every minute" I guess I'm the sucker.

 

[garyeb2]

"There's a sucker born every minute" I guess I'm the sucker.

Only in the context of over valuing content that cost $19.95.

 

[MNsooner]

That's a fair point MN, but unlike football, college basketball has a larger sample size (more games) to formulate more metrics from and incorporate into simulations/rankings. As the season progresses, Vegas will utilize these "multi-simulations" of upcoming games in order to assist in setting their lines. Additionally, you will find more opportunities (no matter how good the team) of teams having "flat spots"....and Vegas knows how to sniff that out and set their lines accordingly.

One last thought regarding basketball/football. You won't find heavy money coming in late/early on a college basketball powerhouse as often as you do on a college football powerhouse because of:
1) Lines aren't usually set until around 24 hours before the game.
2) The margin of error is greatly diminished in college bball because of the larger sample size of data.

I don't doubt that he has a system that is pretty good. I think that kind of stuff is very useful for being able to follow college basketball on the whole especially. I don't have time to watch hundreds of games, but the statistics can give me insight into every team even if I haven't seen them.

That said, if I can actually watch a team every game, I think I can tell you more about that team than the statistics can tell you or at the very least can tell you things the statistics can't.

 

[MNsooner]

Only in the context of over valuing content that cost $19.95.

I don't know how good it is because I've not ever paid to see it. What I do know is that its value is that it is an entertainment product. For the cost of a pair of movie tickets it might be fairly cheap entertainment if you enjoy it.

 

[garyeb2]

I don't know how good it is because I've not ever paid to see it. What I do know is that its value is that it is an entertainment product. For the cost of a pair of movie tickets it might be fairly cheap entertainment if you enjoy it.

Yea, I know. It is like psychology. If you believe, it might do you a little good.

 

[MNsooner]

Yea, I know. It is like psychology. If you believe, it might do you a little good.

:ez-laugh: Exactly.

 

[JDSooners]

I wish I knew the source, I cracked up when I saw it and I think it was an Oklahoma chat post Marcus Smart push with John Helsey, but he said Cobbins was a much greater player than Ryan Spangler, i thought it was crazy and they were totally out of their mind

 

[campbest]

I don't know how good it is because I've not ever paid to see it. What I do know is that its value is that it is an entertainment product. For the cost of a pair of movie tickets it might be fairly cheap entertainment if you enjoy it.

It must be relatively good, every coaching staff in the country uses it. Now that may water down it's effectiveness when the information is readily available to everyone. As Storm mentioned, you don't wait to gameplan the day before a game, those take place over the course of weeks (if you have a staff that's worth a crap).

Not just KenPom, but all the basketball analytics are beginning to scratch the surface. Most of these teams don't use SportVU at the college level, but they will soon enough and then after that new technology will become available and so on. People forget that Pomeroy until two years ago was just doing this in his free time outside his "normal" job. I have it said before and will again, Brad Stevens maintains him and his staff would wait for the KnePom daily updates to directly plug those into his preparations and game day adjustments...he only took Butler to two straight NC games.

 

[bgrch1350]

Cobbins ppg is misleading b/c unless you shoot a jump shot or a 3 pt shot, your not going to score much in Ford's offense. The ball doesn't go through the post players.

 

[oSuJeff97]

I can't really believe that people would argue that losing Cobbins didn't massively effect OSU's team defense.

The evidence is there in KenPom's data, and it's also there if you actually watched all of the games.

It's not like OSU didn't play anybody in the non-conference either.