Questions For The Competition is a weekly column that addresses our issues with the brackets of other bracketology "experts." Today's questions are reserved for ESPN's Joe Lunardi and SI.com's Andy Glockner. Keep in mind that these questions are about each expert's most recent brackets, which were released before Monday's games.Joe Lunardi (ESPN) - Feb. 21 Bracket
What do you have against Richmond? The Spiders are 10-3 in the A-10, they have an OOC win over Purdue, and they're not in? Worse yet, they're behind UAB on your First First Out list?
Speaking of UAB, how are the Blazers so close to being in? Conference USA deserves three bids more than the Colonial does? More than the A-10? Based on what?
How does Cleveland State move up to a 12 seed after getting blown out at Old Dominion?
Isn't Kansas State a little high as a 9? How is Washington still a 7? Virginia Tech is a 10 and is two seed lines above BC?
Andy Glockner (Sports Illustrated) - Feb. 21 Bracket
No Alabama? Memphis and Butler both have better chances at an at-large if the season ended today?
We're not opposed to the Colonial getting three bids, but VCU's an 11? They're not a 12 or Last Four In?
How is UConn two seed lines better than Louisville? The Cardinals beat the Huskies twice and are a game better in conference.
51 comments:
Normally I think you guys are about the best in the business, but you guys really need to get over the whole Alabama thing. I get the fact that you guys look ahead, so I'm assuming you're projecting Bama to finish at least 13-3 or 14-2 in order to include them.
But if the season ended today, they wouldn't be anywhere close to the field. Their #77 in the RPI right now (which would be the worst ever), they're only 3-4 vs. the top 100 and even worse they're only 5-4 vs. the teams between 100-200. They have 2 good wins, and 5 (FIVE!) TERRIBLE losses! How is this an NCAA Tournament team? To answer your question... Butler and Memphis DO have better at-large chances than Alabama if the season ended today.
So what if they're beating up on the pathetic SEC West? That doesn't make up for their 6 non-conference losses, only 1 of which coming against a Tournament team. I think you guys are putting way too much stock into their conference record, and aren't dissecting their actual resume.
Plus what's with you guys assigning random conference maxes and minimums. Teams get bids, not conferences. If George Mason, Old Dominion, VCU, and Drexel were the last 4 at-larges on your S-Curve, would you eliminate 1 of them in favor of a 4th ACC team farther down the S-curve? Just "because the ACC deserves more bids"?
I agree that Memphis, UTEP, and UAB shouldn't all be in Joe's field, but it's not because the CUSA "doesn't deserve 3 bids" it's because UAB and UTEP don't deserve to be in.
^ With all due respect, have you watched Alabama play a game this year after November? Alabama is definitely playing well enough to be tournament worthy. I think their early troubles are being held against them too much.
Tough question to answer, Thomas. It's supposed to be a body of work resume, so, in theory, the games in November would each count as much as games in February.
since when do november games count as much as march games?
B101,
Generally, I agree with everything you say but I have to disagree about the RIchmond/UAB argument. UAB has a significantly better RPI (33) than Richmond (63), as well as 7 RPI top 100 wins vs. 4 for Richmond (granted the Purdue win does look pretty). Also, Richmond has 2 losses outside of the RPI 100 vs. only 1 for UAB.
BTW, I agree with you about Butler. That's an NIT team all the way. As for Alabama...I think the committee might catch hell no matter what they decide.
mag, that's why I said "in theory".
in reality, they don't. a win in the first round of a conference tournament counts A LOT more than a win in november. the "body of work" gives extra weight to more recent games because more recent games are more indicative of how a team will play in the ncaat than games played in november. look at how much a team like UNC has improved during the year.
ESPN has not EVER been kind to A-10. As a Xavier fan it seems that they always say that the A-10 teams will lose when we are up against one of the power conference teams etc. Watch after the selection Sunday show how far their experts advance A-10 teams. It is sad really. I am just happy that Xavier, Temple, Richmond, along with some of the others are fighting for some respect for the conference. If we can get 3-4 teams in a year I would be happy.
Agreed, although the committee claims they quit looking at "last 10" or "last 12" games, last year. In reality, I think you're probably right though, hence Notre Dame's 6 seed, last year.
Joe, last year Hubert Davis picked Butler as his "early exit", if that tells you how much he knows.
I'm curious, does the competition ever actually answer these questions?
We trade tweets with Glockner about his quesions. No response from Lunardi.
Lunardi was actually at UR for the St. Joe's game (he does radio for the Hawks) and saw just how good the Spiders can be.
UR has neutral court wins over Purdue and Wright St and road wins at Dayton and at Seton Hall. Pretty strong.
UAB's best wins are home games over weak opponents. Their RPI is inflated by a bunch of losses to top 50 teams.
UAB is getting consideration because...
1) They are currently 1st in the CUSA standings
2) They have a high RPI now at 30 (one of the highest not included in the field)
3) UAB has just 1 "bad" loss... a 3pt loss on a last second 3pter at Arizona State early in the year.
4) UAB has 7 (SEVEN) Top 100 wins
5) UAB has played the last several games without their leading scorer/rebounded Cameron Moore (including a 4pt loss at Memphis)... Cameron Moore will be back later in the year. The committee has shown it makes special consideration for injuries.
Alot of the "hate" on UAB is directed at the absence of a Top 50 win... yet several of UAB's victories are just outside the Top 50 and have the potential to become Top 50 wins.
UAB has wins over the following teams (including some which have been listed as bubble teams)...
- VCU (RPI 57)
- Marshall [twice] (RPI 59)
- UTEP (RPI 61)
If all of those become Top 50 RPI teams, that would be 4 Top 50 wins. In any event, there's a shot that some of those teams reach the Top 50. UAB also has a remaining game @ Southern Miss (RPI 44) that is yet another shot at a Top 50 win.
Lots of at-large teams with worse resumes than UAB are getting more consideration. And it has almost everything to do with the perception that CUSA is weak (despite 7 teams from CUSA being in the RPI Top 100). CAA, by the way, has just 6 teams in the RPI Top 100.
If Syracuse wins @Gtown on Sat, do they climb to a 3 seed? They have great road wins (St Johns, UConn, and Nova) along with home wins over WVU and ND. Compare them to Purdue or Wisconsin. Those 2 teams beat OSU at home recently, but not much else is in there, especially on the road.
Anonymous I remember well, Butler is my wife's alma mater and we watch a lot of their games since she has a few relatives that played for them. I actually was telling her about that same thing while we were sitting up in Indy getting ready to walk over to Lucas Oil to watch them play for the national championship.
Secretly though I was wishing Xavier had knocked off K-State because I REALLY wanted another shot against the Bulldogs after the clock mess up earlier in the season last year.
Syracuse will be a 3 if they win at Georgetown this weekend.
RPI is overrated. In the RPI, a 40 point loss to a great team is more valuable than a 40 point win over a terrible team.
B101,
How high can Michigan State climb if they win out the regular season and make it to the conference semis?
We know that teams get bids and not conferences, but when people make that statement they seem to forget the human element that is involved. The committee members are not robots and whether or not they want to admit it each of them will be thinking for example "If we put Marquette in that will be 11 Big East teams."
We are admitably down on UAB more so than other bracketologists since they have already been swept by Memphis and will likely be swept by Southern Miss after their game in Hattiesburg on March 2. That will put the Blazers behind both Memphis and Southern Miss in the C-USA pecking order which likely won't be a a good place to be. Their tourney hopes are definitely improved with Memphis recent lose to Rice and UTEP losing to UCF in the past week, but the third best team out of C-USA will likely equal NIT this year.
Michigan State could climb to a 7 given your scenario
@Joe, I like the A-10 too, but some team besides Xavier needs to do damage in the NCAAs for the committee to take the conference seriously.
You know what guys who projected early exits for all of the A-10 teams in last year's tourney, behind their backs, should be called? Correct, 66.66 percent of the time. 12 teams in the league, 1 sweet sixteen berth... Gosh, How can they possibly refrain from showing every game on ESPN? 1 team out of 12 made it to the sweet sixteen the year before, as well... And, that's weird... 1 the year before that. The A-10 is a 1-team league that gets pulled up because of Xavier. Take Xavier out and you might as well be the patriot league.
In fact...
Since 2005's tournament, the Patriot league has 2 wins in the tourney. The A-10, minus Xavier, has 2.
Lawyered.
Will, I agree the RPI is overrated. When will the selection committee chuck it altogether and adopt a more accurate ratings system?
UAB is currently 1st in CUSA.
If UAB finishes the season with wins over Houston, Southern Miss, and ECU... UAB will win at least a share of the CUSA regular season title (if not outright). Do that, and you can take it to the bank. UAB will be in the NCAA tournament.
Don't like the RPI?
One of the other more popular rating systems was developed by Sagarin which factors in margin of victory.
Here are the Sagarin ratings for some teams which have been discussed:
UAB (52)
Butler (53)
Gonzaga (54)
Alabama (58)
Boston College (59)
Colorado State (62)
Memphis (67)
Baylor (68)
Those are the synthesized Sagarin ratings. I'm pretty sure the committee uses the ELO chess rating (margin of victory does not matter). The predictor rating uses margin of victory, but again I don't think the NCAA uses that (BCS only uses ELO).
Looking at the ELO...
Memphis (43)
UAB (48)
Boston College (51)
Colorado State (54)
Tennessee (55)
Clemson (56)
Butler (60)
Gonzaga (66)
Alabama (79)
Baylor (82)
I dont think they use Sargarin nearly at all.
In fact its listed as "other reference materials" so the numbers (along with Kenpom) are available for use but from what I've heard people hardly ever use them.
margin of victory should not be used because there are too many other factors that it doesn't account for. for example, why should a team that goes up by 30 in the 2nd half get penalized for emptying its bench and letting the backups play if the final margin of victory winds up only being 15 points?
Jerry Palm now projecting UAB in along with Memphis from CUSA.
Palm's Last 4 in...
- Richmond
- Michigan
- VCU
- Gonzaga
Palm's Last 4 out...
- Dayton
- Drexel
- Marquette
- Southern Miss
The NCAA using any ratings system that included point spreads would only enourage teams to run up the score.
Clemson should be getting more consideration. They have beat 4 teams in Joe Lundardi field. FSU,Long Beach St.,BC and VT.
Margin of victory should certainly be included.
In fact, studies have shown that margin of victory is far more indicative of true talent then actual W/L record.
To protect from what others were saying you can put a cap on it so that teams aren't encouraged to "run it up".
If I have two teams and one is 25-10 and the other is 20-15 but the 25 win team is barely winning games meanwhile the 20 win team is blowing everyone out, I'm taking the 20 win team all things being equal.
To just ignore margin of victory your leaving out a huge portion of data.
"In fact, studies have shown that margin of victory is far more indicative of true talent then actual W/L record."
give me a break. one, what "studies"? two, who cares about talent? so should we just seed based on who has the most mcd AAs on their rosters?
margin of victory is so flawed it's ridiculous. as i wrote earlier, there are so many variables that do not get accounted for (eg, playing the bench in a blowout). also, how would you account for bc losing by 32 to unc AT HOME and then going on the road and losing to the same unc team (no major injuries) by 2 on the road 18 days later?
I'm waiting to hear how a team with 15 losses is blowing everyone out.
You misunderstood.
1. By true talent I simply mean the better team (not necessarily the ones with the most mcds).
2. There's been studies on this in EVERY SPORT. In baseball you could look at runs scored and runs allowed as a better indictor of future performance than W/L record. This is the very same thing in basketball. I'll see if I can find a link later.
3. The 15 was just an example. Obviously an NCAA team with 15 losses isn't blowing out many people so maybe my numbers I used were bad in these circumstances.
So your trying to tell me that if Team A won by 25 over Team B and Team C won by 5 over Team B, you'd treat each as the same?
Thats ridiculous. But that's what your doing by totallying excluding margin of victory.
Now obviously u get to a point of diminishing returns. A 20 point victory is far better then 10 point but is a 40 point win that much more impressive than say a 30 point one? So eventually you get to a point where it matters less and less which is why I suppose you could cap it (just so that teams don't purposely run it up).
In fact, look at the NCAA winners the past 10 years...look at their margin of victory....compare it to other teams. It most certainly means alot.
http://www.nytimes.com/2008/03/17/sports/ncaabasketball/17score.html?_r=1&scp=6&sq=NCAA%20RPI%20North&st=cse
Theres one.
One last thought,
You need to consider strength of schedule obviously. A 30 pt win over Depaul isn't the same as a 30pt win over Pittsburgh lol.
Actually this is exactly what Ken Pomeroy does with his rankings. I'm oversimpliying it but basically its margin of victory but adjusting it for SOS.
But it's cool that people think margin of victory doesnt matter. It just means I make more and more money when I do my brackets :)
Also your thing about playing the bench shouldn't matter a great deal. First off, how many times do you see the bench play significant minutes? Even in blowouts I rarely see teams empty the bench with significant time remaining. Yes so a 25 pt lead might turn into a 15 pt win.....that's still better then a 5 pt win.
The great majority of teams unload their benches at the same time during blowouts so IMO the effect is small on the overall season results.
I agree that margin of victory can be used as an aid to rate teams. The question is whether or not it ought to be.
Predicting future results isn't what's important in selecting tournament teams. What's important is evaluating past performance. And ultimately, all that matters in that evaluation process is who you play and whether you win or lose.
I wouldn't want some widely inconsistent team to get a bid over a more consistent team, simply because the former had more blowout wins against the same schedule.
If two teams play the exact same schedule, I'll take the one with more wins, every time.
If two teams play the same schedule, I could care less which one had a better point differential or which one beat more "top 50" teams. All I care about is which one had the better record.
I agree to an extent but margin of victory isn't just about predicting the future.
I mean it all depends on what your wanting. The teams with the most wins might not be the better team.
IMO it's about sending the BEST 37 or so at large teams.
I'm just not so certain that just because one team has more wins against the same schedule, that they are the better team. I hate to say this but its true with all sports......luck does play a role in things....that's why I hate looking at wins/losses only.
I think you look at as much data as possible and then make the decision (w/l, margin of victory, rpi, kenpom, sargarin, who you beat, where you beat them etc.)
Glad to see the Horizon debate about Butler is still raging in here...!!
The RPI is by far the WORST model out there at measuring how good teams are. If you honestly believe differently, then this article will explain why you are wrong better than I can:
http://www.bigbluehistory.net/bb/rpi.html
the study in the article you posted really is a bunch of bunk. they looked at ncaa basketball games over 4 seasons in which teams played each other twice. i assume that most of such pairings were home/away conference games. its conclusion was that because the team that wins the first game isn't more likely to win the second game, there is no such thing as a "clutch" team. it also found that if a team wins by 20 the first game, it has a greater than 50% chance of winning the 2nd game. this isn't very surprising because (1) the home court is an enormous advantage so the home team usually will win in conference play and (2) if one team is much better than the other, it doesn't matter where they play. it definitely doesn't prove that margin of victory indicates which team is better.
"In fact, look at the NCAA winners the past 10 years...look at their margin of victory....compare it to other teams. It most certainly means alot."
it means a lot if you are comparing ohio st to iowa. it doesn't mean anything when you are comparing ohio st to kansas.
are you surprised that the ncaa champions have better margins of victory than most other teams? do they have the best margins of victory? no.
i beg to differ about teams rarely pissing away huge leads. yes, you don't see a team empty its bench with 10 minutes to go in a blow, but you usually see them liberally rotating in bench guys and resting starters. you also see a team with a huge lead often playing stall ball or practicing different offenses and defenses, which causes the lead to dwindle.
It's not just comparing Ohio St to Iowa. Anyone can see the differences there without even considering margin of victory.
Read the first part. About not picking UNC and going with Kansas because UNC had alot of close wins as opposed to Kansas. Sure enough Kansas won the title that year. Now obviously UNC was a great team and they still could have won the title but in terms of who the safer bet was they went with the team that had the better margin of victory. And thats one example.
This isn't really about determining who's better between 1 and 68 but rather 1 and 2 and 2 and 3. Obviously there's more things to look at. I would never suggest just looking at this and concluding one team is better then the other. But to completely ignore data is wrong.
It's not surprising that others feel this way. Thats why teams get underrated and overrated all the time.
In order to get a truly accurate rating system, margin of victory MUST be included. I would suggest putting an upper limit of 20 points (no more credit for wins by more than 20 points).
I don't think margin of victory should matter, at least not that much.
I know this is a minimal sample, but look at Butler in the tourney last year--they beat UTEP by 18, but then over their next five games (including the Duke loss) the margin of victory wasn't any greater than seven. The point is, they still GOT THE WIN.
Then a team like, say, Kentucky, who blew out everyone in their first three games, loses when they play a close one against West Virginia.
Obviously that's just a random example that probably doesn't carry much weight, but I'd almost argue having more close wins is better, since that's a more battle-tested team.
Tough query Diablo 3 itemsto resolve, Johnson. It's supposed to be a body of work resume, so, the theory is that, the game titles throughout Nov would likely every count just
Diablo 3 Gold kaufenas much as games in March.
[p]Yet, continually exert understanding with regards to an internet site . This [url=http://www.fashionhermesbagsuk.co.uk]discount hermes bags kelly[/url] kind of bag is a well-known name within the industry of designer bags . If this suit isn't going to fits completely, you should not fret about it, I have ability to cope with the make a difference is only the Herve Leger gown which can be a bit of cake . Most of Hermes bags Cheap replica handbags have been renamed immediately after the popular icons of Hollywood's like Kelly and Birkin . types and types in varied inductive leather at EuroHandbag truly are a open spell using eyes . Coach Outlet Bags Gucci Outlet Presently, gals rather have selecting persons totes which might be streamlined, hip yet still efficient along with in which,Coupons For Coach Outlet Replacement Move bags include the most suitable alternative . Victoria appeared with a hermes handbag birkin in New York
Just as we said before that there could not be such a place as good as New York and Paris for you to meet top stars and [url=http://www.springhermeshandbags.co.uk]hermes handbags birkin[/url] celebrity . Hermes loved types is divided into three parts
Hermes loved types is divided into three parts
Today, Hermes loved types is divided into three parts: Hermes Sellier; La Montre Hermes; and Hermes Parfums . Refer to it's origin, there is a little story: one day of 1984, while the famous celebity Jane Birkin seated next to Hermes CEO [url=http://www.springhermeshandbags.co.uk]hermes handbags sale
[/url] Jean-Louis Dumas on a Paris to London flight, she opened her Hermes-made datebook and a flurry of loose notes fell to the floor.[/p][p]This creates these bags so unique and distinctive brand name of bags within the market area that potential customers wait for these bags with abated breath . Hermes has many masterpieces . In 1961, she wears a white hat and gloves and a green Givenchy dress, with a Kelly Bag to return to the United States, its elegant charming style to try reading charms are not telling secretly Kennedy, a skirt around his fashion dialogue, suddenly bring fashion and round of "Grace agitation" . They are available in a number of different leathers and skins, which is another reason for the beginning price tag of $7000 . Super sized favorites like the Hermes Birkin or Chanel Denim Cabas announce to the world that you take your purse needs very seriously and are willing to pay for it . Thanks to modest sizes, [url=http://www.fashionhermesbagsuk.co.uk]hermes bags kelly[/url] Kelly bag constantly produces exclusive tasteful impression . A massive a component relating to functioning moreover other [url=http://www.fashionhermesbagsuk.co.uk]hermes belts women
[/url] commonly effective communication . But whatever [url=http://www.springhermeshandbags.co.uk]hermes handbags kelly uk[/url] the reason is, it looks as if the couple茅聢楼忙陋職 relation was hardly affected.[/p]
Post a Comment