|
Post by neutralfan on Dec 29, 2013 13:31:57 GMT -6
Kevin Thanks for all the good work, and you showed, in my opinion, that we are trying to fix something that is not broken. May need tweaking, but not an overhaul like is being proposed. Also another thing that is being overlooked is that last year the coaches did not truly understand how to "play the system" but as the understanding grows so to will the number of undeserving teams making the playoffs based solely on power rankings. Finally, surely a system is not right when Mandeville ends up with fewer power points than Hammond, after they won by a landslide, is both stupid and unfair. Thanks This argument has been going on for nearly a decade now... Back when Flat3 first devised the soccer power rating that would work best with LHSAA. We have DISCUSSED it, used it, tweaked it, and studied it for years now. I was always a proponent of coaches seeding. There are problems with it... But all in all it does a good job of seeding. The biggest problem we face is having teams qualifying that should not be in the playoffs over teams that should! If that doesn't mean your current system is BROKEN, then what does? Teams that played around less than 10 games last year and won 3 or 4 getting in over a team like Opelousas Catholic who would have been somewhere around 16 or 17 in power ratings for Division III! Any playoff system that omits a team that might have even hosted a playoff game in favor of a team that the top 12 teams JV could defeat is not a fair playoff system. power ratings has been asked of us for years now... And the time has come to step up and be a sport like the rest of every other LHSAA sport. Report our schedules, play our matches, report our scores! No more sitting at home and only reporting when you win games to laprep. No more canceling your games cause you don't feel like going... There are cancellation policies in the LHSAA once they take over with power ratings! This sport has grown by leaps and bounds over the past few years and its time to be a part of the LHSAA system like EVERY other team sport. I'd rather see a team get a seed too high or too low (it happens now with seeding anyway, go ask Northlake Christian about last year) than a team that absolutely should be in the playoffs not make it. A team that finished fourth in a district that had 3 top 10 teams that could win most other districts in their division... The time for too much emphasis on district play is over in other sports and power ratings reward the district champs only... Then the strongest team fill the bracket! Once a seasons worth of games are played... Power ratings are usually spot on. Sure, they look funny at times during the season or after a game in particular.... But once a season is played out... It is usually spot on and the teams that deserve to be in the playoffs will be in it. once it is in... We will have the LHSAA official records being posted for ALL teams! We will know exactly what EVeRY team is doing! It will be like every other team sport and we will finally stop being the one team sport that the LHSAA doesn't have info on nor knows what is going on... And most importantly, the right teams will all get in the playoffs! Neutralfan... what example of power point are you using with the last statement with Mandeville and Hammond? We have not released any power ratings this year nor in any other year with the finally agreed upon formula that uses a mix of flat3 and LHSAA regular formula. This isn't a sudden proposal... This has been studied and tested for years now. And today we will roll out the first example of the proposed agreed formula to show what is happening in regards to if this season would already be using it... And once the end of the season data is completed, it should be a great comparison to the current model in place. The example I was using is based on the fact Mandeville will get about 22 wins and Hammond probably 0. Mandeville gets 10 for win 0 for Hammonds wins. Hammond 0 for loss 11 for 50% of Mandeville's wins. Unless I have the math wrong, so I am open to corrections. Thanks
|
|
p_malinich
Data Expert
www.elevenlions.com
Posts: 4,201
|
Post by p_malinich on Dec 29, 2013 14:09:04 GMT -6
The example I was using is based on the fact Mandeville will get about 22 wins and Hammond probably 0. Mandeville gets 10 for win 0 for Hammonds wins. Hammond 0 for loss 11 for 50% of Mandeville's wins. Unless I have the math wrong, so I am open to corrections. Thanks Your math is right. To keep it neutral, let me switch to Team A who finishes the year with 22 wins and Team B who finishes the year with 0 wins. Let's assume that they are district opponents and MUST play each other twice during the season. In that scenario, the PR points earned for those games definitely benefits Team B. In this year's proposal, when you win, you get 5 points (not 10 as in previous proposals) plus 100% of the wins (at end of season) of the other team. In a losing effort, you get 0 points plus 50% of the other team's wins. So Team A (the better team) earns a Power Rating of 5.000 (5 for win & 0 for opponent) for those 2 games. Team B (the lesser team) earns a Power Rating of 11.000 for those 2 games. But that's only for those 2 games. The reality is that their are many other games that work into the averages of both teams. That's why BCcoach references that there are oddities during the year. That's why we didn't publish PR in November - generates more irregularities than anything. So what's the impact of the math?Team A - They must play Team B twice since they're in District, but I doubt they will schedule many (if any) other opponents like Team B during the rest of the season. No blowout wins to improve their record. They may need to face teams like that in tourneys, but again, I expect that they manage schedule to get better games (and in theory, help their own team improve more). Team B - They probably don't like getting destroyed. They, too will play their required District opponents, but probably won't schedule just those games. They'll also want games in which they can be competitive. In the end, the averages should work themselves out. I'm sure a few may still try to manipulate the system. They may make it to the playoffs, but will probably exit first round, regardless of seeding method.
|
|
|
Post by lashowcase on Dec 29, 2013 17:15:01 GMT -6
I appreciate and respect everyone’s opinions on the subject of the power ratings proposal, whether they are in favor of the proposal or not. We proposed power rating as an objective way to determine and seed teams for the playoffs. I am well aware that this new system has advantages and disadvantages, but it is consistent and it forces everyone to play by the same rules. I coach a team that has benefited from being seeded very high during the coaches seeding process over the last four years, so we did not propose power ratings for any type of personal interest. The only reason that I have been an advocate for power ratings over the last few years is because I feel like the sport that I love has been compromised. I have had numerous discussions with coaches about how coaches have made deals with other coaches. These deals include ranking certain teams higher and intentionally ranking other teams lower. Some of these discussions were with coaches that have admitted to brokering deals with other coaches or have admitted to seeding teams lower because of a personal grudge against another team’s coach or team.
Ultimately, the reason why basketball went to power ratings was because of the same manipulation that was occurring during the coaches’ seeding process.
If this proposal doesn’t pass, than hopefully it make coaches think twice about manipulating the process. If this proposal is passed, hopefully it will only add to the high expectations and integrity that we all expect from our sport.
Jason Oertling St. Louis Catholic Soccer
|
|
|
Post by neutralfan on Dec 29, 2013 17:16:48 GMT -6
Hi p_malinich Thanks for the correction don't know what I was thinking, to much holiday cheer. I still have concerns though (..In the end, the averages should work themselves out. I'm sure a few may still try to manipulate the system. They may make it to the playoffs, but will probably exit first round, regardless of seeding method. ..) I agree that they SHOULD work themselves out but there are the cases of teams just wanting to make the playoffs. The reverse of that is the teams like Mandeville and SSA who are going to be adversely effected by having to play teams twice who have few if any wins. While they are obviously going to make the playoffs it could mean, as someone has pointed out, that they may have to travel in the quarter finals. So go ahead use the power rankings during regular season but let the coaches seed. Again just my 1c worth
|
|
|
Post by d4enthusiast on Dec 29, 2013 17:19:02 GMT -6
I agree to Boomer's points, but I have more to add: 1) Teams do not have a choice on what district they are in. Why is it fair to play in a district that usually means 2 games per team and those teams are terrible? They won't have decent records and the better teams are 'penalized' for playoff position by winning games against teams that won't pull wins against anyone. Not fair. I don't buy into this. Every single district in the state currently has a team with 2 or fewer wins. This seems to affect everyone equally.
|
|
p_malinich
Data Expert
www.elevenlions.com
Posts: 4,201
|
Post by p_malinich on Dec 29, 2013 18:13:54 GMT -6
They are published. It's a 3rd tab with Standings, Schedules & Results, and Power Ratings. You can link from that main thread, but here's a direct link to the Power Ratings* Team, Division, District - pretty self-explanatory * Power Rating - The actual calculation - it can change even when you don't play since it's based on your opponent's wins as well * Rank Position - What the team's position is when sorted across all divisions in Power Rating order * OT Wins & OT Losses - How many games went to OT & resulted in other than a tie * PR Wins, PR Losses, & PR Ties - Adjusted standings to reflect any game post regulation as a tie * PR Win # - The team's win # for PR purposes (PR wins + 1/2 of PR ties) Top 5 teams as of results reported through this a.m. 1) St. Paul's (13.929) 2) Catholic BR (11.625) 3) Brother Martin (11.500) 4) Bolton (11.288) 5) Northlake Christian (11.117) You can view all 150+ teams at the link. You can filter for just one division (or any criteria). Feel free to ask questions about the calculations. I'm sure they're out there. I'll do my best to explain them. Enjoy and let the discussions begin (or continue)...
|
|
|
Post by methuselah on Dec 29, 2013 18:48:11 GMT -6
A cursory review of D-II seems pretty solid.
|
|
|
Post by Boomer on Dec 29, 2013 19:47:05 GMT -6
A cursory review of D-II seems pretty solid.
Yes, it does look OK ... for D-II... for a specific mathematical reason that there are fewer D-II programs and the hierarchy is pretty clear in any case. In D-II, the computer is not materially different from the educated coaches poll. However....
Just look at the mess in D-I and D-III and imagine seeding the State playoffs with that ranking? Do you really believe Jesuit should be ranked #12 in D-I? Jesuit vs. Lafayette at Lafayette in the second round?
The computer ranking of D-III is a joke. Who on this green earth thinks Lusher is the 16th ranked team in D-III, ranked behind Ascension Episcopal among others? Have you seen Lusher play?
Example 2: Who thinks Dunham et. al. is better than Newman? If that is the case, so is St. Charles Catholic and Haynes, both of whom were tied by Dunham this weekend in Baton Rouge ... (Note: no intent to disparage Dunham, comparison only used for illustrative purpose... )
Saying "we've been talking about it for ten years" is no argument for adopting a flawed and destructive system that will never be undone once done. The reason "we've been talking about it for ten years" is that every computer system we have tried stinks. And the mathematics tend to create a skewed geographical model favoring teams from one section of the State, probably because of circular reasoning.
This formula, and all the simplistic formulas tested to date produce these kind of head scratching results. And using them to seed the brackets will be a lot worse than continuing to coach seed. This will not end well.
|
|
|
Post by beauchenecoach on Dec 29, 2013 20:05:54 GMT -6
I can only laugh at goes who can not understand why certain teams power ratings are where they are... Just go look at their schedule! It's why you have these numbers to see how it is effected before the end if this year. Don't schedule teams that will only win 2 or 3 games. Already you will have that in your district, so why schedule a bunch of non district games against teams that will not win 3 or 4 games? Are you really worthy of being a top 10 team with a schedule like that? Why? Cause a human being was impressed with that 10-0 win? I understand your concerns and agree to a point... But when the season is all said and done, your wins will count and who they were against will matter! Those teams on your schedule need to win around 10-15 games for you to have a top 5-10 power rating! Plain and simple!
And don't give me this DII is different! It isn't different! Just cause there isn't a ton of NoLA schools in DII doesn't make it different! The formula doesn't take a stance that NOLA or Acadiana area is better than another area and maybe that is best! And if there was a flawed division with coaches seeding, it was definitely DIII last year! And I agree with Coach Oertling (who is Division III by the way!), there were way too many under the table agreements... And even some who brag about it. Some who claim an east vs west thought process or a north vs south, or a BR area vs NOLA area.... Whatever the case, each year has had more and more complaints from coaches about seeding and the deals. The reason these power ratings are put is to show you how your scheduling effects the outcome! Instead of just seeing a team lower than its coaches poll ranking and complaining... Do a little investigating into each teams schedule and you can instantly see why they are low. Now with all that said... The big controversies with power ratings in the future will not be from a team too low but from teams too high! That can be manipulated with some geographical dominance over a few teams and all of them winning a large amount if games against weaker teams and thus creating an anomaly that makes sense in the numbers but not from objectivity. That is my one concern once coaches figure out how to schedule. Either way... We will get the right teams into the playoffs this way and first and foremost,that should be the biggest goal. Getting the quarter finalist seeded correctly should be second... Anything behind that is up to each team to get hot and play their best in February!
|
|
|
Post by mswatd on Dec 29, 2013 20:16:36 GMT -6
I can only laugh at goes who can not understand why certain teams power ratings are where they are... Just go look at their schedule! It's why you have these numbers to see how it is effected before the end if this year. Don't schedule teams that will only win 2 or 3 games. Already you will have that in your district, so why schedule a bunch of non district games against teams that will not win 3 or 4 games? Are you really worthy of being a top 10 team with a schedule like that? Why? Cause a human being was impressed with that 10-0 win? I understand your concerns and agree to a point... But when the season is all said and done, your wins will count and who they were against will matter! Those teams on your schedule need to win around 10-15 games for you to have a top 5-10 power rating! Plain and simple! And don't give me this DII is different! It isn't different! Just cause there isn't a ton of NoLA schools in DII doesn't make it different! The formula doesn't take a stance that NOLA or Acadiana area is better than another area and maybe that is best! And if there was a flawed division with coaches seeding, it was definitely DIII last year! And I agree with Coach Oertling (who is Division III by the way!), there were way too many under the table agreements... And even some who brag about it. Some who claim an east vs west thought process or a north vs south, or a BR area vs NOLA area.... Whatever the case, each year has had more and more complaints from coaches about seeding and the deals. The reason these power ratings are put is to show you how your scheduling effects the outcome! Instead of just seeing a team lower than its coaches poll ranking and complaining... Do a little investigating into each teams schedule and you can instantly see why they are low. Now with all that said... The big controversies with power ratings in the future will not be from a team too low but from teams too high! That can be manipulated with some geographical dominance over a few teams and all of them winning a large amount if games against weaker teams and thus creating an anomaly that makes sense in the numbers but not from objectivity. That is my one concern once coaches figure out how to schedule. Either way... We will get the right teams into the playoffs this way and first and foremost,that should be the biggest goal. Getting the quarter finalist seeded correctly should be second... Anything behind that is up to each team to get hot and play their best in February! I'm following quite nicely now with all the Q & A happening. I do agree that a power ranking system encourages scheduling tougher opponents to give your team better standings - and I like that! I'd rather a loss to a #3 ranked DI team than an easy win over a #5 ranked DIII. It keeps the big boys playing one another and gives those on the cusp an opportunity to schedule and fight for better PR. Right?
|
|
|
Post by Boomer on Dec 29, 2013 20:23:22 GMT -6
The big controversies with power ratings in the future will not be from a team too low but from teams too high! That can be manipulated with some geographical dominance over a few teams and all of them winning a large amount if games against weaker teams and thus creating an anomaly that makes sense in the numbers but not from objectivity. ... Getting the quarter finalist seeded correctly should be second... Anything behind that is up to each team to get hot and play their best in February! Chad you and I see the desired result the same way, but perhaps come to different conclusions about the cure.
Perhaps there was some deal making last year, if so shame on those people. But I think the anomalous seeding in D-III last year was caused by publishing the unofficial computer ratings. Too many lower case D-III voting teams just adopted the seriously flawed computer rankings. The polls prior to the seeding, votes by conscientious and informed voting coaches, were pretty good in my opinion.
But if the skewed D-III seeding last year was caused by the influence of a bad computer ranking, why try to cure it by mandating adopting a bad computer ranking? Surely there is an alternative.
And here is my idea. Large districts, seven-eight teams per district, carefully balanced for strength by a elite group of knowledgable senior people. Ignore the travel requirements because teams are traveling anyway. Now every team has a mandatory 12-14 games played in district against a mix of team strength. No team will be slacking, filling up their schedule with cheapos while only playing a few required district games. And because of the number of district games, no one will be penalized because of not-enough-games.
If the districts are formed so that each has two-three prospective strong teams, two-three also-rans, and three middle-of-pack, even if some teams prove better and some worse than expected, the results should be comparable. 3 teams from each district qualify eight districts mean 24 teams qualify with eight wild cards.
Advantage... districts and district titles are meaningful, all teams have an automatic scheduled 14 games, computers not needed.
|
|
|
Post by gobbo on Dec 29, 2013 20:39:09 GMT -6
Personally, I like the power rankings, they seem to be very accurate by year end, but if you continue to have the coaches poll to determine seeding, why not publish each coaches ranking? Maybe they would think twice about making deals if they knew their poll was to be published.
|
|
|
Post by methuselah on Dec 29, 2013 20:44:32 GMT -6
With regard to Dunham -having watched our team play them twice this year, I can attest that they are a very quality side. They are technically sound and have very good soccer players and athletes. I don't think their results at the tournament this weekend was wholly indicative of their season long strength. The word I heard at Burbank this weekend was that they had some key contributors missing due to holiday travels or something (of course you play who you brung and if their results suffered it is what it is - but I wouldn't use it to make a judgment on a long term item like power rankings).
With regard to the rankings encouraging teams to play tougher - I think the top tier teams learned this lesson long ago - it pays great dividend to the improvement of your team to play the top teams. Amd becaise these teams have had such good long term reputations, it didnt' unduly hurt them to lose some games to these other top teams.
I think it's been the 2nd tier teams that may have at times been hesitant to play a lot of top teams for apprehension that a loss or two here would knock them so low in the rankings it would effect things like home field and opponent in the playoff seedings. With power rankings, I think you'll see more and more of these 2nd tier teams looking to play top teams because there is something to be gained even in a loss or a tie in the form of points. This can only help competitiveness in the long run in my opinion.
|
|
usfgk
Data Expert
Posts: 495
|
Post by usfgk on Dec 29, 2013 20:56:45 GMT -6
With regard to Dunham -having watched our team play them twice this year, I can attest that they are a very quality side. They are technically sound and have very good soccer players and athletes. I don't think their results at the tournament this weekend was wholly indicative of their season long strength. The word I heard at Burbank this weekend was that they had some key contributors missing due to holiday travels or something (of course you play who you brung and if their results suffered it is what it is - but I wouldn't use it to make a judgment on a long term item like power rankings). With regard to the rankings encouraging teams to play tougher - I think the top tier teams learned this lesson long ago - it pays great dividend to the improvement of your team to play the top teams. Amd becaise these teams have had such good long term reputations, it didnt' unduly hurt them to lose some games to these other top teams. I think it's been the 2nd tier teams that may have at times been hesitant to play a lot of top teams for apprehension that a loss or two here would knock them so low in the rankings it would effect things like home field and opponent in the playoff seedings. With power rankings, I think you'll see more and more of these 2nd tier teams looking to play top teams because there is something to be gained even in a loss or a tie in the form of points. This can only help competitiveness in the long run in my opinion. Here lies the problem. There is no incentive for the top teams to play 2nd tier teams. Top teams will, out of necessity, only schedule other top teams or continue the trend of playing out of state quality opponents instead of playing traditional or local teams that won't positively influence their power ratings. I think that scheduling will become contentious in the future under this system.
|
|
|
Post by Boomer on Dec 29, 2013 21:02:04 GMT -6
... I'd rather a loss to a #3 ranked DI team than an easy win over a #5 ranked DIII. It keeps the big boys playing one another and gives those on the cusp an opportunity to schedule and fight for better PR. Right? If your team would lose to the #3 D-I team, which is Brother Martin in both polls and PRs, but can get an "easy win" over the #5 D-III team, which is either University (PRs) or Lusher (polls), you have an interesting perspective on comparative strengths of these teams. I'm not sure how many teams will get an "easy win" against U or Lusher PLUS... you get a lot more PR points for beating University or Lusher than you do for losing to BM.
Note: By no means was I disparaging Dunham... they are a quality side. As I said... it was for illustrative reasons only ... but there are some other curious results, at least from my perspective, in D-III PRs.
Not sure if this is true. We have lost three games this year.... to highly ranked, higher D teams Mandeville, East Jefferson, Grace King, each loss was by a two goal margin. How did those tough losses help Haynes' PR? We would have been better off beating a couple of D-III coed teams who have a positive record playing each other. How did earning a draw with Rummel (who tied Vandebilt etc.) help our PR?
I totally agree, and think it is already contentious. I believe the answer lies in big, carefully balanced districts that require a mix of district games against good, medium, and trailing teams.
|
|
|
Post by mswatd on Dec 29, 2013 21:12:09 GMT -6
I'm tracking now! And by no means did I intend to disrespect any DIII teams with my example. Just trying to get a grasp on the situation. Apologies if anyone is upset with my example.
|
|
|
Post by methuselah on Dec 29, 2013 21:15:24 GMT -6
Here lies the problem. There is no incentive for the top teams to play 2nd tier teams. Top teams will, out of necessity, only schedule other top teams or continue the trend of playing out of state quality opponents instead of playing traditional or local teams that won't positively influence their power ratings. I think that scheduling will become contentious in the future under this system. I have to respectfully disagree. If you look at the "Also Receiving Votes" and the bottom portion of the top 10's, these teams tend to have a significant number of wins. Which I think would be attractive to a top team from a power points perspective.
|
|
|
Post by kevin on Dec 29, 2013 21:18:45 GMT -6
Either way... We will get the right teams into the playoffs this way and first and foremost,that should be the biggest goal. Getting the quarter finalist seeded correctly should be second... Anything behind that is up to each team to get hot and play their best in February! I think those are both worthy goals, but I think I might rank them the other way. Let's say a team barely missed out on the playoffs due to its division record, but would've made it in as a #28 seed in the power ratings. Are they likely to beat the #5 seed? No. Now, let's say a team ranked #2 in the polls, widely considered top-tier, decides to play an out-of-state tournament and a couple games against Mississippi or Texas schools. Maybe their district was a little weak. Their ranking according to the power ratings is #9. They easily win their first two matches, then they host the #1 seed in the quarterfinals. To me, the issue of seeding the best teams is probably more important to the playoffs as a whole. The teams that are barely sneaking into the playoffs (regardless of the method used) are probably going to get knocked out in the first round anyway. Here's how I see the pros and cons of the power rating system: Pros: more deserving teams will make the playoffs better regular season scheduling (fewer teams trying to dodge competitive same-division opponents) Cons: seedings could be way off a region could end up with unfairly inflated rankings I want to stress that I haven't made my mind up about this. I don't know what I'm going to tell my principal. (Right now, I'd lean slightly in favor of the proposal, just because I don't think the seeding will be too much different but that better teams will make the playoffs and we'll have better regular season matches.) I ran the numbers in an attempt to figure out how the seedings would work. For the most part, I don't think it will make much difference. What I worry about is the "black swan" situation that pops up every now and then and ruins a bracket (like the #2 team getting seeded as #9). I also think that, given a few years time, there will be teams that try to manipulate the power ratings. (Not that I blame them for trying; nor do I blame a team in the current system that tries to go out and schedule the worst same-division opponents it can in order to make the playoffs.) I think that, on the whole, power ratings would provide a better field (I listed the teams that would've been in and out in my earlier post; if anyone disagrees, feel free to let me know). But I think that the current system of seeding works well (the numbers bear that out), regardless of the reports of collusion to rank teams unfairly. If that's a problem, let's figure out how to address it. If the deal-making and collusion is so rampant that the system is completely broken, then fine, let's get rid of it. But I haven't seen anyone make that case.
|
|
usfgk
Data Expert
Posts: 495
|
Post by usfgk on Dec 29, 2013 21:24:29 GMT -6
Here lies the problem. There is no incentive for the top teams to play 2nd tier teams. Top teams will, out of necessity, only schedule other top teams or continue the trend of playing out of state quality opponents instead of playing traditional or local teams that won't positively influence their power ratings. I think that scheduling will become contentious in the future under this system. I have to respectfully disagree. If you look at the "Also Receiving Votes" and the bottom portion of the top 10's, these teams tend to have a significant number of wins. Which I think would be attractive to a top team from a power points perspective. We disagree on definitions. I would not consider a top ten or top fifteen team as 2nd tier.
|
|
|
Post by Boomer on Dec 29, 2013 21:56:52 GMT -6
The computer generated state PR ranks all the soccer teams against each other, ignoring divisions, etc. If the PR system has any credibility, it ought to generate a State-wide ranking that is somewhat believable. It doesn't ... and that is the same complaint the basketball people have.
Here are the current comparative PR top twenty teams ... lets be sure we understand... the computer thinks this is the proper strength ranking of all the teams in the State. And if we had a playoff bracket that ignored divisions, Jesuit would travel to Menard for their first round game.
1. St. Pauls 2. Catholic BR 3. Brother Martin 4. Bolton 5. Northlake Christian 6. Acadiana 7. St. Louis 8. Lafayette 9. Sulphur 10. Northshore 11. Mandeville 12. Beau Chene 13. Grace King 14. Captain Shreve 15. Menard 16. Westminster 17. St. Amant 18. Jesuit 19. University 20. Destrehan.
|
|