The Things that Matter for Good Teams

[DobbsRover2]

In all honesty I did not have the time to read your epic post and I was being somewhat playfully flippant BUT....

Here's what he actually said (ESPN quote) "I've been coaching 41 years and this has never happened," DePaul coach Doug Bruno said. "This doesn't happen in sixth and seventh grade. It doesn't happen. It happened. I trust we have the inner strength to turn the page."

I sure the rest of your post was more fact-based. Numbers are really all that matters.

If you want to unquestioningly take Doug's word about that, that's your right. I'm sure his teams have never blown that many FTs against ND in 2014 before, as that's kind of a one-off situation.

His teams have been historically bad FT shooting teams, though they excel in other areas. Two years ago they lost to Dayton at home in an early season loss by caroming 10 of their 18 FTs, and later that year they allowed a terrible 9-21 Pittsburgh squad to hang around till the end by going 7-16 in a 55-53 squeaker - again, AT HOME.

News to Doug, stinkers very much like the ND one are going to come your way again in the future, because when your teams live constantly in the overall 64% to 68% range at the FT line, fecal stuff happens. Then again, maybe you should just play all your games on the road.

 

[Bayareahusky]

Yeah i couldnt believe it watching him go through the agony.... he even tried to "ice" his own free throw shooter so that fortunes may change......

Whether ND won or lost doesn't matter to me..... What I feel badly about is the lift a DePaul win would have given to his program....... but as he said, they are going to have to find the inner strength to overcome

 

[UcMiami]

Yeah i couldnt believe it watching him go through the agony.... he even tried to "ice" his own free throw shooter so that fortunes may change.

Whether ND won or lost doesn't matter to me..... What I feel badly about is the lift a DePaul win would have given to his program.. but as he said, they are going to have to find the inner strength to overcome

Stuff happens. Same thing with Duke loss and Uconn loss. As they say - it builds character!

 

[KnightBridgeAZ]

Inclusion of "scoring margin" in the original post seemed a little off to me. Clearly among a group of teams with good offense and relatively good defense, scoring margin should be good. But it can also be good for a much weaker team that plays and dominates much weaker competition, or even a good team that beats the crap out of the little sisters but struggles in 5 or 6 games a year against better competition, albeit winning most of them by a close margin.

And, using a different phrasing that I think folks have spoken to, there is a difference between the "macro" and "micro" scale; free throw shooting can be very important in an individual game or situation that it might not reflect over-all. Similarly with some of the other stats.

I think good assist / turnover margin, good shooting, good rebounding and decent defense will do the trick, regardless of stat specifics.

 

[UcMiami]

The problem with any of these stats is that the major factor in each one for very good teams is determined by the 80+% of their games played against inferior teams - e.g. scoring margin ... what we do against UCD and CofC bumps our averages up in so many categories that it could hide issues and inferior stats in games against the few Stanfords and NDs we play.
It can just as easily pump up the numbers for teams at the other end of the spectrum - Green Bay for example is top five in wins over the last X years - they would be no where near that if they were an ACC team - they would still do quite well, but not in the same conversation with the NDs and Uconns of the D1 world.

 

[DobbsRover2]

Inclusion of "scoring margin" in the original post seemed a little off to me. Clearly among a group of teams with good offense and relatively good defense, scoring margin should be good. But it can also be good for a much weaker team that plays and dominates much weaker competition, or even a good team that beats the crap out of the little sisters but struggles in 5 or 6 games a year against better competition, albeit winning most of them by a close margin.

And, using a different phrasing that I think folks have spoken to, there is a difference between the "macro" and "micro" scale; free throw shooting can be very important in an individual game or situation that it might not reflect over-all. Similarly with some of the other stats.

I think good assist / turnover margin, good shooting, good rebounding and decent defense will do the trick, regardless of stat specifics.

Not sure what the problem with including scoring margin is, though obviously you would very much expect a top team among the majors to have a good scoring margin unless it was winning a lot of close ones and maybe had a few blowout losses if a team has to play the #1 team three times. The worst ranking among the Elite 8 teams was Texas A&M at #29 with an 11.0 margin. All of the teams ahead of that mark were either good majors or mid majors except Arkansas, which played a truly awful OOC schedule with all but as home games and they ran up the margin. It is possible to be one of the top teams in the country with a fairly low scoring margin, but it would be unusual, and that is why it is in the stat list. All stats are affected by a team's SOS in many ways, but yes, that's the story for any sport.

And yes, near the top of the list for scoring margin are a bunch of the top mid majors, but again, what's the problem? They are good teams for whom scoring margin is as important as it is for a major team, and yes they play a lesser schedule but they still need to dominate their opponents. If I had run the stats for the top 8 mid major teams like Gonzaga and Dayton, I would expect the stat rankings relevance to be similar to the Elite 8 teams.

As with any stat there can be unusual cases. Western Illinois had a +5.7 scoring margin but a losing record, while USC Upstate was at -4.1 and had a winning record. But with many of the 17 stats used here, you can rank near the bottom of one (Texas A&M at #339 in 3pt FGs per game) and still be a top team if your FG% is at #19 and your FG% defense is #28 or other important categories. But being outside of the top 50 teams in Scoring margin and being one of the top 8 teams would be unusual, and that is why it shows up as the most relevant stat.

 

[DobbsRover2]

The problem with any of these stats is that the major factor in each one for very good teams is determined by the 80+% of their games played against inferior teams - e.g. scoring margin ... what we do against UCD and CofC bumps our averages up in so many categories that it could hide issues and inferior stats in games against the few Stanfords and NDs we play.
It can just as easily pump up the numbers for teams at the other end of the spectrum - Green Bay for example is top five in wins over the last X years - they would be no where near that if they were an ACC team - they would still do quite well, but not in the same conversation with the NDs and Uconns of the D1 world.

That does not have a great deal of relevance to the stats computed for teams across the whole season, and statistics tend to average out over a season, so even if UConn shoots FTs poorly against say ND, they will also likely have good days against some other top team.

All stats in any sport are affected by SOS, but unless you want to be a stat-Luddite you accept that along with the law of averages. Do we toss out the Raptors' stats this year because they play a lot of turkeys like the Knicks? Some of the mediocre stats you get against good teams are balanced by the good stats you get against mediocre teams, and the things you do well and not so good show through.

We know that UConn does not go 100% for 100% of the game with their best players against SMU, but their stats for most of the things they do well should be pretty consistent. And the stats may may not be quite as glossy for the Huskies against good teams, but they scored 76.7 ppg against ranked opponents last year, only about 5.4 points under their overall average, so the for them and other teams, the stats to a large extent even out.

And whether or not Green Bay's stats would be less glossy if it played in the ACC has no real relation to which stats matter for the top major teams, and a couple of mid majors' effects on stats rankings is fairly trivial, as they are just a couple of the 344 teams. Once you start picking on the schedule of a team like Green Bay, then you start tumbling down the slippery slope alongside all the major teams like Arkansas and Indiana who play poor schedules with worse overall SOS than many of the top mid majors.

 

[UcMiami]

That does not have a great deal of relevance to the stats computed for teams across the whole season, and statistics tend to average out over a season, so even if UConn shoots FTs poorly against say ND, they will also likely have good days against some other top team.

All stats in any sport are affected by SOS, but unless you want to be a stat-Luddite you accept that along with the law of averages. Do we toss out the Raptors' stats this year because they play a lot of turkeys like the Knicks? Some of the mediocre stats you get against good teams are balanced by the good stats you get against mediocre teams, and the things you do well and not so good show through.

We know that UConn does not go 100% for 100% of the game with their best players against SMU, but their stats for most of the things they do well should be pretty consistent. And the stats may may not be quite as glossy for the Huskies against good teams, but they scored 76.7 ppg against ranked opponents last year, only about 5.4 points under their overall average, so the for them and other teams, the stats to a large extent even out.

And whether or not Green Bay's stats would be less glossy if it played in the ACC has no real relation to which stats matter for the top major teams, and a couple of mid majors' effects on stats rankings is fairly trivial, as they are just a couple of the 344 teams. Once you start picking on the schedule of a team like Green Bay, then you start tumbling down the slippery slope alongside all the major teams like Arkansas and Indiana who play poor schedules with worse overall SOS than many of the top mid majors.

The issue is that if 30 out of 40 games (and for Uconn it is sometimes 35 out of 40) games are uncompetitive from around the 10 minute mark of the first half, Geno starts substituting in bench players for major minutes - assists, shooting percentage, rebounding, turnovers, etc. all have a tendency to suffer and by 10 minutes of the second half defensive pressure, fast breaks, and starters are seldom seen. To put it the other way - for those maybe 15 minutes when Uconn is playing full out against a CofC the stats are all stunning because they are shooting uncontested or nearly so. If 3/4 of the input into a data set is relatively meaningless, the output is less than definitive. ND had the highest scoring offense prior to playing Uconn and that didn't exactly prove anything when they could only score 58. The fact that they scored 105, 104, and 112 against UML, HC, and Quin (those NE power houses) really was meaningless and if they had really tried they probably could have increased both their scoring and their MOV in those games beyond the 63 point average they did achieve. Add in the Uconn loss to those three and they have an MOV of 42 - very impressive but it still only gets you a 3-1 record.

 

[baylorprof]

I love you guys' board, but this thread is making my head hurt. Maybe this is the kind of "cipherin" it takes to win 10 national titles.

So while you're at it, help this Baylor academic understand why we "shoot" a "throw" and call a "throw" a "shot?" Knowing an answer might give us the edge we need over Texas, and lordy, lordy we'll need all we can muster to hang with those big girls.

 

[UcMiami]

I love you guys' board, but this thread is making my head hurt. Maybe this is the kind of "cipherin" it takes to win 10 national titles.

So while you're at it, help this Baylor academic understand why we "shoot" a "throw" and call a "throw" a "shot?" Knowing an answer might give us the edge we need over Texas, and lordy, lordy we'll need all we can muster to hang with those big girls.

Simple rule of thumb: Good teams shoot, bad teams throw!
That should be a very interesting game! Good luck to you all.

And Dobbs - I just think in really competitive games for any of the top 10 teams the Stats based on the YTD do not have much bearing on how the game plays out because most of the data in those stats is built off blow-outs against inferior teams. You can speculate that teams will have good assist/TO of good shooting percentages but against quality defense that may not come to pass. Uconn had 14 assists and 24 turnovers against ND - not close to their YTD A/TO average of 1.3. And ND shot 31% not close to their YTD average of 49%.

 

[Zorro]

One last point, and then I promise to shut up about free throws (shots). It is undoubtedly true that some players' performances improve in pressure situations, and that other players' perform worse. The former are able to use the pressure to better focus their attention and ignore outside factors, etc., while the latter's muscles may tense with the apprehension of failure, etc. (This is the point of calling a time out before a crucial FT or field goal try by the opponent/) However, no amount of practicing in the gym is going to alter that. A player whose psychology causes him/her to tense up in tight situations is going to do so no matter how many thousands of practice shots he or she has taken.

 

[DobbsRover2]

Simple rule of thumb: Good teams shoot, bad teams throw!
That should be a very interesting game! Good luck to you all.

And Dobbs - I just think in really competitive games for any of the top 10 teams the Stats based on the YTD do not have much bearing on how the game plays out because most of the data in those stats is built off blow-outs against inferior teams. You can speculate that teams will have good assist/TO of good shooting percentages but against quality defense that may not come to pass. Uconn had 14 assists and 24 turnovers against ND - not close to their YTD A/TO average of 1.3. And ND shot 31% not close to their YTD average of 49%.

Sorry, citing a a stat or two from one game doesn't make for much of a case, just as it doesn't for me saying that Huskies pad their stats in big games like the NC last year by grabbing 12 more rebounds than their season's average or making 4 more assists than usual, or in Saturday's game by getting 52 rebounds and 9 steals. That would also be wrong and foolish.

The major components of a team's stats do not come from a few blowouts but from the vast weight of all the games on the schedule, where both good and bad performances even out. Almost all of UConn's games were blowouts last year, and there will be swings in stat numbers like 6 blocks against Memphis followed the next game by 17 against Houston, but in the end they averaged 8.1 bpg, a number they got in 8 games last year and they also got between 6 and 10 in 30 of 40 games. One big game with a stat against a weak team does not affect the average much, and note that UConn got 8 or more blocks in 8 games against the theoretically tougher ranked opponents, so the stat wasn't based just on getting the 17 against Houston or the 4 against UC Davis in less taxing games.

Again, unless we want to base everything on "hunches" and "perceptions," which I know is the more popular way to go, the stats can provide general trends and statuses for those who want to look at them as a whole. Yes, in every game we can point to a stat that was very different from a team's larger season's average, but in stats one game does not a season make nor does 24 turnovers or 17 blocks make a season's stat average.

 

[DobbsRover2]

One last point, and then I promise to shut up about free throws (shots). It is undoubtedly true that some players' performances improve in pressure situations, and that other players' perform worse. The former are able to use the pressure to better focus their attention and ignore outside factors, etc., while the latter's muscles may tense with the apprehension of failure, etc. (This is the point of calling a time out before a crucial FT or field goal try by the opponent/) However, no amount of practicing in the gym is going to alter that. A player whose psychology causes him/her to tense up in tight situations is going to do so no matter how many thousands of practice shots he or she has taken.

Truly, the head is likely the most important part behind shooting FTs well, or poorly. And when you are Karl Malone at the line and sneaky Scottie Pippin is whispering in your ear, "The Mailman doesn't deliver on Sundays," the pressure can be at the boiling point. I never had to shoot FTs with hundreds of distractions be waved by a background crowd chanting, "Miss. Miss," but it's got to be tough to get focused, especially late when your arms and legs are made of lead.

 

[DobbsRover2]

The issue is that if 30 out of 40 games (and for Uconn it is sometimes 35 out of 40) games are uncompetitive from around the 10 minute mark of the first half, Geno starts substituting in bench players for major minutes - assists, shooting percentage, rebounding, turnovers, etc. all have a tendency to suffer and by 10 minutes of the second half defensive pressure, fast breaks, and starters are seldom seen. To put it the other way - for those maybe 15 minutes when Uconn is playing full out against a CofC the stats are all stunning because they are shooting uncontested or nearly so. If 3/4 of the input into a data set is relatively meaningless, the output is less than definitive. ND had the highest scoring offense prior to playing Uconn and that didn't exactly prove anything when they could only score 58. The fact that they scored 105, 104, and 112 against UML, HC, and Quin (those NE power houses) really was meaningless and if they had really tried they probably could have increased both their scoring and their MOV in those games beyond the 63 point average they did achieve. Add in the Uconn loss to those three and they have an MOV of 42 - very impressive but it still only gets you a 3-1 record.

Again for one who deals in statistics, the fact that you are trying to build a case off of a handful of early season games against a few of ND's weak opponents doesn't mean much to me, and just as preseason rankings often don't mean a whole lot, and that what happens to a team's offense when it goes up against the defensive team in the nation is not necessarily a strong indicator of what will happen against the other 38 or so opponents, the point with stats is to look at the bigger picture that does not get skewed by a few unusual results. I also don't remember any games last year where UConn's top 5 only played 15 minutes, even in the worst of blowouts like with Monmouth, and maybe basing arguments on starters getting 25 or more minutes would be more accurate.

Once again, using the big picture of a whole season, ND did have the second highest scoring offense last season at 86 and was the best shooting team at 50%, and they played against the best defense in the country in the NC game and the figures were 58 and 35%, way down from their usual average but still well above the average for UConn's opponents. But to say that ND simply just built up their scoring average against the weakest teams is wrong. Going into the NC game last year, ND had played 13 games against ranked teams and averaged 85.3 points. Agreed, that's 0.8 below their season's average, but for anyone to say that ND's scoring offense stat was based only on beating up on patsies is not right. As to what you want to make of a few early season games and a few isolated stats, sure, whatever.

 

[UcMiami]

Dobbs - last year for example (40 games) Uconn had 9 games against top 10 teams and 31 against all others.
Against top 10 teams their scoring average was 74.7 against all others 84.2 and because their were 31 games against the all others universe their full year number was 82.1.
Against top 10 the opponents averaged 56.2 PPG, all others 45.4, and total was 47.8
Against top 10 MOV was 18.4, all others 38.8, and total was 34.3

Those are pretty divergent numbers over the course of the year and they would probably reflected in the other stats as well. They would diverge even more did not Uconn put the breaks on their offense and call off the pressure defense and liberally use the bench players in many of the 31 games against non-top ten teams. That is all I am pointing out - when 3/4 of the games in a season are hopeless mismatches the statistical analysis suffers. Uconn treats every possession seriously at both ends, but the team does not go all out to score as many points as it can and limited the opponents scoring as much as it can in most of the games it plays in any year. And the same is true to varying degrees with all of the top teams each year. Gene would much rather work on the team's zone defense, or specific concepts on offense, or get bench players a chance to learn in live action, than to pound a team into submission once the contest is decided. (Syracuse upon occasion excepted!)

 

[DobbsRover2]

Sure, there are going to be some differences in stats when teams play a top 10 team versus when they play a UC Davis, and sure your starters won't be putting in 36 minutes in a blowout, whether it' against a weak team like SMU or an Elite 8 team like Louisville.

We both know that we can pick and choose and slice up stats until we find ones that fit our point. But I'm not sure I completely get the point about comparing the nine top 10 opponents against the rest. Like any team in the WCBB, UConn's schedule did not include a solid row of top 10 teams. Everybody plays a mixed bag schedule, and their performance and stats are based on many different scenarios. And if I was to say that most of ND's stats overall last year were worse against the top team last year, well that's a no-brainer that's hardly worth mentioning.

So why choose an unrepresentative group like the top 10 teams to make a comparison when you can choose a group that's closer to UConn's real average of the schedule? Take the five games against ranked opponents not in the Top 10. Against those teams UConn's margin was 29.0 points based on the 80.0 - 51.0 average score. That's not too far off the overall season's average, somewhat better than against the top 10 teams, worse than against the next batch of teams in the top 75, and still worse than the teams rated worse than 100. But that is the nature of the schedule for all teams, and until UConn starts playing 40 games against top 10 teams or 40 games against non top 10 teams, determining a true season's average for them from one or the other is not that useful in the real world. And like the pro teams that have stats from games against the Spurs and the Knicks, you have to go with what is in play.

 

[cabbie191]

I love you guys' board, but this thread is making my head hurt. Maybe this is the kind of "cipherin" it takes to win 10 national titles.

So while you're at it, help this Baylor academic understand why we "shoot" a "throw" and call a "throw" a "shot?" Knowing an answer might give us the edge we need over Texas, and lordy, lordy we'll need all we can muster to hang with those big girls.

Hey Prof - seems like a very long time since you posted anything. Welcome back!

 

[cabbie191]

My last thoughts on the subject. In basketball, at least in so far as this thread is going, the stats are really interesting but ultimately what we will care most about are the tournament results at the end of the year. Yes, stats may be useful as predictors of overall performance but they won't make a difference in any individual game once the first tip-off takes place.

I'm going to shift focus slightly. As a Wisconsin resident who follows Badger football, my emotional side hopes that Gordon gets the Heisman Trophy. But even before the Ohio State game, I didn't think he would be a likely winner because I sense that voters often go more for hype and don't really analyze the stats.

Everyone (at least to hear it from the television broadcasters) is gaga about Mariota and all the touchdowns he has thrown this year. But if you look closely at the statistics, you'll find that there are four Pac 12 quarterbacks with 30 or more TD passes this season, and two more with 27 and 28 respectively. Mariota is a great quarterback, but when you look at the conference he plays in, I think some of the shine comes off. Four of the ten worst pass defense teams in the country are in the Pac 12.

 

[MilfordHusky]

I'd argue that missed FTs cost us 1 or 2 of the 7 losses to ND. I think we missed 3 or4 in overtime of one of the losses. I think we also missed 1 late in the 2012 Final Four game.

 

[DobbsRover2]

My last thoughts on the subject. In basketball, at least in so far as this thread is going, the stats are really interesting but ultimately what we will care most about are the tournament results at the end of the year. Yes, stats may be useful as predictors of overall performance but they won't make a difference in any individual game once the first tip-off takes place.

I'm going to shift focus slightly. As a Wisconsin resident who follows Badger football, my emotional side hopes that Gordon gets the Heisman Trophy. But even before the Ohio State game, I didn't think he would be a likely winner because I sense that voters often go more for hype and don't really analyze the stats.

Everyone (at least to hear it from the television broadcasters) is gaga about Mariota and all the touchdowns he has thrown this year. But if you look closely at the statistics, you'll find that there are four Pac 12 quarterbacks with 30 or more TD passes this season, and two more with 27 and 28 respectively. Mariota is a great quarterback, but when you look at the conference he plays in, I think some of the shine comes off. Four of the ten worst pass defense teams in the country are in the Pac 12.

Look, I'm a Badger fan too since my dad was a 1942 alumni, but when your team gets blasted 59-0 in their last regular season game, you can kiss the Heisman good-bye even if you run for 666 ypg and score 4TDs every quarter on average. I love my stats, but when something like 59-0 happens, I chuck'em. Just like with Baylor two years ago, you may be the classiest pink caddy on the road to the NC, but if you go down 82-81 in the Sweet 16, you're only a rambling wreck for that season.

 

[BigBird]

And of course they must have been way better than almost any college team last year, as less than 7% even hit that 75% minimum standard for her team. College players just don't put in the practice I guess.

That's what more than one coach has told me. The team I was affiliated with was expected to practice FT's outside of the team's plenary practice time. Some did it a little, some did more. And one or two shot hundreds of FT's.