maligned Posted February 5, 2014 Share Posted February 5, 2014 Here are the Duals State Championship qualifying scores for 1A through STATE (no changes from semi-state). Teams earning automatic positions will receive their official invites shortly after scores are made official. The voting for the final two positions will most likely be in early to mid-April. Teams eligible for those positions will be sent a standardized form in which to include information about JV or 8th grade talent and/or individuals returning from injury that will affect their quality for next season. **Scores are UNOFFICIAL until FRIDAY. (If you believe a record-keeping error was made, please send me a pm immediately.) Clinched Semi-State Representative or Automatic Wildcard: Churubusco Union County Bremen Tecumseh West Central Monrovia Central Noble Milan Westview Triton Central Clinched Vote-in Consideration (clinched eligibility for both vote-in positions unless specified otherwise): Prairie Heights Eastside Fremont Tell City Adams Central Seeger Winchester North White South Putnam South Adams North Knox LaVille (eligible for final vote-in position only) **Scores are still unofficial [table] Class 1A Teams Total Underclassmen to State Points Available Churubusco 330.78 0 0 Union County 289.86 0 0 Bremen 286.2 0 0 Tecumseh 285.86 0 0 West Central 280.32 1 4 Monrovia 273.36 0 0 Central Noble 268.35 0 0 Milan 267.44 0 0 Westview 265.05 0 0 Triton Central 263.56 0 0 Prairie Heights 260.72 0 0 Eastside 258.56 0 0 Fremont 256.35 0 0 Tell City 255.54 0 0 Adams Central 254.74 1 4 Seeger 251.03 0 0 Winchester 248.88 0 0 North White 246.18 0 0 South Putnam 246.1 0 0 South Adams 244.43 0 0 North Knox 236.64 0 0 LaVille 229.34 0 0 Woodlan 226.32 0 0 Lake Station Edison 224.94 0 0 Fountain Central 219.96 0 0 Frankton 219.31 0 0 Elwood 218.08 0 0 Winamac 215.57 0 0 Eastern Hancock 214.16 0 0 Lewis Cass 213.48 0 0 Clinton Central 212.64 0 0 Triton 211.92 0 0 Knightstown 206.75 0 0 Wood Memorial 201.68 0 0 Shenandoah 200.56 0 0 Rossville 199.24 0 0 Cloverdale 199.14 0 0 Eastern (Greentown) 192.26 0 0 River Forest 190.44 0 0 Monroe Central 188.92 0 0 Southern Wells 180.88 0 0 Hagerstown 177.76 0 0 Tri 177.74 0 0 Culver Community 172.87 0 0 Lapel 172.84 0 0 South Newton 167.15 0 0 Providence 166.29 0 0 Northeastern 166 0 0 Indianapolis Scecina 164 0 0 Southwood 162.8 0 0 Clinton Prairie 162.78 0 0 Indianapolis Broad Ripple 162.24 0 0 Covington 162.21 0 0 Daleville 161.12 0 0 Wes-Del 159.86 0 0 Southwestern (Hanover) 158.76 0 0 Manchester 157.66 1 4 Union City 155.96 0 0 Hammond Bishop Noll 150.16 0 0 New Washington 149.98 1 4 Forest Park 147.04 0 0 Turkey Run 146.56 0 0 North Judson 146.11 0 0 Hebron 145.26 0 0 Eastern (Pekin) 143.39 0 0 North Vermillion 140.44 0 0 Carroll (Flora) 139.82 0 0 Attica 138.76 0 0 Indianapolis Howe Academy 134.59 0 0 North Newton 134.16 0 0 Frontier 132.8 0 0 Indianapolis Crispus Attucks 131.89 0 0 Tri Central 129.41 0 0 Cascade 129.34 0 0 North Miami 128.2 0 0 Indiana School for the Deaf 126.6 0 0 Cambridge City Lincoln 124.79 0 0 Wapahani 121.62 0 0 Taylor 121.04 0 0 Northfield 119.2 0 0 Whiting 115.86 0 0 Faith Christian 114.6 0 0 Caston 111.96 0 0 Alexandria 111.89 0 0 Wabash 108 0 0 Sheridan 100.86 0 0 Crawford County 100.58 0 0 Switzerland County 97.78 0 0 Madison Grant 97.68 0 0 Paoli 96.52 0 0 Bluffton 95.04 0 0 Indianapolis Manual 94.7 0 0 Park Tudor 86.02 0 0 Rockville 83.64 0 0 Pioneer 81.41 0 0 Speedway 80.11 0 0 Blue River Valley 77.39 0 0 South Spencer 58.64 0 0 Christian Academy 42.77 0 0 Randolph Southern 31.64 0 0 Lafayette Central Catholic 19.2 0 0 Howe School 18.24 0 0 Madison Shawe 17.8 0 0 Indiana School for the Blind 15.4 0 0 Oldenburg Academy 0 0 0 0 0 0 0 [/table] Link to comment Share on other sites More sharing options...
maligned Posted February 5, 2014 Author Share Posted February 5, 2014 *Especially Merrillville semi-state teams should be considered unofficial, as they have not been cross-checked for accuracy Link to comment Share on other sites More sharing options...
maligned Posted February 5, 2014 Author Share Posted February 5, 2014 One more thing: I'll be on an IndianaMat Google hangout tomorrow night (probably 8pm...not announced yet officially) if you want to ask any and all questions related to Duals State Finals Qualifying or Genius ratings I'll address any questions posted here or any questions you bring up during the hangout Link to comment Share on other sites More sharing options...
rookie78 Posted February 6, 2014 Share Posted February 6, 2014 The second number listed is the number of remaining underclassmen; the third number is the total remaining wrestlers I am assuming the second number is the # of underclassmen going to regionals and the third # is total number of wrestlers remaining and advancing to regionals, not the number of returning kids for next year Link to comment Share on other sites More sharing options...
maligned Posted February 6, 2014 Author Share Posted February 6, 2014 The second number listed is the number of remaining underclassmen; the third number is the total remaining wrestlers I am assuming the second number is the # of underclassmen going to regionals and the third # is total number of wrestlers remaining and advancing to regionals, not the number of returning kids for next year Right...underclassmen still alive (they score significantly more points) and total wrestlers still alive Link to comment Share on other sites More sharing options...
maligned Posted February 12, 2014 Author Share Posted February 12, 2014 Now updated through regional Link to comment Share on other sites More sharing options...
bog190 Posted February 12, 2014 Share Posted February 12, 2014 What would teams with low multiplier have to do to overcome Churubusco's post sectional score (329)? If I have it figured correctly, a realistic score for AC is about 262, that's after the full tournament, with 2 state qualifiers (one senior and one underclassmen), and an underclassmen ticket rounder, as well as 2 underclassmen first rounders, Adams Central has 2 seniors this year, so they return the majority of their team. Churubusco had 329 with 6 regional qualifiers (3 underclassmen and 3 seniors). They also have 5 seniors total. So, based off of these scores, AC's full tournament, with 5 total semi state qualifiers, 4 of them underclassmen, and 2 state qualifiers, one of them underclassmen, is worth 67 less points than Churubusco's 6 regional qualifiers. I understand AC looks like they will be at minimum in voting range, but my question is, what would they have to do to outscore Churubusco's post sectional score? Take everyone to Semi State? Have 4 state qualifiers? Link to comment Share on other sites More sharing options...
maligned Posted February 12, 2014 Author Share Posted February 12, 2014 What would teams with low multiplier have to do to overcome Churubusco's post sectional score (329)? If I have it figured correctly, a realistic score for AC is about 262, that's after the full tournament, with 2 state qualifiers (one senior and one underclassmen), and an underclassmen ticket rounder, as well as 2 underclassmen first rounders, Adams Central has 2 seniors this year, so they return the majority of their team. Churubusco had 329 with 6 regional qualifiers (3 underclassmen and 3 seniors). They also have 5 seniors total. So, based off of these scores, AC's full tournament, with 5 total semi state qualifiers, 4 of them underclassmen, and 2 state qualifiers, one of them underclassmen, is worth 67 less points than Churubusco's 6 regional qualifiers. I understand AC looks like they will be at minimum in voting range, but my question is, what would they have to do to outscore Churubusco's post sectional score? Take everyone to Semi State? Have 4 state qualifiers? Look, I'll be the first to admit that 1A is by far the most difficult class to develop the scoring system for. The turnover is very unpredictable and all the forfeits and such make it tough to predict duals quality. I spent hours in the weeks leading up to this tournament analyzing all angles for how we could make it as accurate as possible. In the end, I'm more satisfied with how it is now that how it was before. Had we used it last year, we would have had all the same strong teams that were at the event (AC, Busco, Fremont, South Put), but we would have also caught Cass and North White and Union County. Do I think Busco will be much better than AC next year--no (although the final gap is probably going to be more like 325 to 265, rather than how it appears now). Do I think AC will really be about the 9th or 10th best team, rather than top 3--no. But I think we're getting an accurate group of strong teams and we'll have the chance to analyze deeply a nice group of teams after the top 10 to pick a couple of successful additional teams, as we did last year. Link to comment Share on other sites More sharing options...
bog190 Posted February 12, 2014 Share Posted February 12, 2014 Look, I'll be the first to admit that 1A is by far the most difficult class to develop the scoring system for. The turnover is very unpredictable and all the forfeits and such make it tough to predict duals quality. I spent hours in the weeks leading up to this tournament analyzing all angles for how we could make it as accurate as possible. In the end, I'm more satisfied with how it is now that how it was before. Had we used it last year, we would have had all the same strong teams that were at the event (AC, Busco, Fremont, South Put), but we would have also caught Cass and North White and Union County. Do I think Busco will be much better than AC next year--no (although the final gap is probably going to be more like 325 to 265, rather than how it appears now). Do I think AC will really be about the 9th or 10th best team, rather than top 3--no. But I think we're getting an accurate group of strong teams and we'll have the chance to analyze deeply a nice group of teams after the top 10 to pick a couple of successful additional teams, as we did last year. The way the system is this year allows Churubusco to be invited in almost any circumstance because of their multiplier. They literally could have packed it up after sectional and not have been challenged the rest of the way. If AC loses a few tough matches at semi-state they end up down by huge margins. Link to comment Share on other sites More sharing options...
maligned Posted February 13, 2014 Author Share Posted February 13, 2014 Whether we'd done the old system, this system, or any system that considers path difficulty at all, a team in 1A with all of its guys but one placing top 6 in a difficult sectional (and most advancing to higher levels) and with a healthy complement of guys returning is going to qualify easily. Even with no multiplier applied to anyone, 'Busco is top 10--but they've gotten those advancement levels in a much more crowded path than the other teams that would be with them in the top 10. I do agree that the final slotting of AC will be a little lower than ideal (though they'll still get in the event), but it's more about the fact that we've still not managed to capture some of the year-forward negative impact in 1A of having 2-3 forfeits (and the comparative benefit of a full lineup) as well as I'd like. It has nothing to do with Busco. Link to comment Share on other sites More sharing options...
bog190 Posted February 13, 2014 Share Posted February 13, 2014 Whether we'd done the old system, this system, or any system that considers path difficulty at all, a team in 1A with all of its guys but one placing top 6 in a difficult sectional (and most advancing to higher levels) and with a healthy complement of guys returning is going to qualify easily. Even with no multiplier applied to anyone, 'Busco is top 10--but they've gotten those advancement levels in a much more crowded path than the other teams that would be with them in the top 10. I do agree that the final slotting of AC will be a little lower than ideal (though they'll still get in the event), but it's more about the fact that we've still not managed to capture some of the year-forward negative impact in 1A of having 2-3 forfeits (and the comparative benefit of a full lineup) as well as I'd like. It has nothing to do with Busco. I understand that it's not about Busco, they are just the clearest example. You can say the same about Eastside, who will score similar to AC because they fill most of the lineup and go to Carroll. I'm still interested in what AC would have to do to have a similar score. Link to comment Share on other sites More sharing options...
infowrestling Posted February 13, 2014 Share Posted February 13, 2014 I think that this shows that the gap in the multiplier is just far too wide......at least it is for 1A. Maybe we should have separate multipliers for each class and sectional. I know that I've heard several wrestlers and regular fans complaining that this process is just far far too simple and easy to understand.....let's add some more complexity to it. Link to comment Share on other sites More sharing options...
bog190 Posted February 13, 2014 Share Posted February 13, 2014 I think that this shows that the gap in the multiplier is just far too wide......at least it is for 1A. Maybe we should have separate multipliers for each class and sectional. I know that I've heard several wrestlers and regular fans complaining that this process is just far far too simple and easy to understand.....let's add some more complexity to it. This is my whole point. Eastside had 7 wrestlers not place in the top 6, yet they sit at 258, a number AC will likely surpass by a small margin, but may not surpass if they lose a match they shouldn't. Eastside had 7 sectional qualifiers, 4 regional qualifiers, and 1 semi state qualifier. AC will have to get 2 state qualifiers to match that. It's out of whack. Link to comment Share on other sites More sharing options...
littlevito Posted February 13, 2014 Share Posted February 13, 2014 If you check out the 2A thread, it gives better examples of how the multipliers work. The clarity is a lot better. Link to comment Share on other sites More sharing options...
bog190 Posted February 13, 2014 Share Posted February 13, 2014 If you check out the 2A thread, it gives better examples of how the multipliers work. The clarity is a lot better. I understand how the multipliers work. Does it sound like I'm asking how they work? I'm arguing that the multipliers are out of whack, especially in 1A. Link to comment Share on other sites More sharing options...
maligned Posted February 13, 2014 Author Share Posted February 13, 2014 I'm arguing that the multipliers are out of whack, especially in 1A. I explained this to someone else. They are actually very accurate--on average. But as someone made a sarcastic remark about a few posts ago, they're not accurate in every single situation. They work in a general sense, but in some isolated instances some teams will be able to take advantage of a place where the multiplier helps them that they wouldn't have done any better anyway (or vice versa). They would work better if we applied different multipliers for each class at each level of the tournament--but then we might as well just go to a full-on, full-season, computer-generated model. But there's still a desire to have a "sporting" method where teams can score points and track their progress. As I've said with 1A, there really is an element with teams having forfeits that I've only figured out how to apply to the model I'm tinkering with in the last couple weeks. This would have moved AC up some and left a few of the higher teams back more. This to me--having the multiplier and the 1A issue of small lineups working "in concert"--is the most concerning point. In the end, we're going to end up with strong teams as the auto bids, and we'll have some good candidates for the voted-in spots (and the chance to leave behind some teams with slightly lesser resumes). And the three year collective run of data we've had has still brought more than 75% year-end Top Ten teams for the 1A event. But 1A will continue to be a work in progress until we feel there are some better ways to hone in on teams' year-forward talent even more accurately. Link to comment Share on other sites More sharing options...
AC Computer Posted February 13, 2014 Share Posted February 13, 2014 I think the multiplier range is too large, also, but don't have the math background to justify it. I would think a +/- .15 would be more fair, but that's just personal opinion, not any kind of calculation. Part of my opinion is that the multiplier is based on the Genius Rating for the current year. As I understand them its a method similar to that used to predict point spreads for college and/or professional contests. I'm guessing its similar to the Sagarin ratings for HS football. I'm not sure that wrestling lends itself to this system very well, though. First, not everyone reports all of their dual results. I try to post every one of AC's results if someone else hasn't already done so, but I see very few results for a lot of other class A teams. How does that affect the Genius Ratings? If a team only reports its wins will it help their rating? I also think that wrestling does not lend itself to this type pf prediction model as basketball or football. In those sports a coach can change his line up, but the team still plays together as a unit to achieve victory. In wrestling its 14 different individual matches that determine the result and a line-up change, like substituting a JV wrestlers to take a forfeit and moving your wrestler up against a very good wrestler from another team can have a greater effect than a line change in the other sports. I think that's why in wrestling more than any other sport you can't compare the team results against common opponents and predict results. The other thing I'm interested in is the points awarded to 5th/6th place winners at sectional. I agree that these wrestlers should be awarded more points than the entry that eliminated immediately and that its more important to a class A team than class 3A, but I think that the difference for class A and 3A teams between SEC5/6 and Reg is too small. The difference for 2A may be too large. A 2A underclassman that advances to the regional scores 13 more points than one that finishes 5th, but 3A only scores 4 more points and 1A adds only 2 points. I think this gives a little too much credit to the 5th/6th place finishers in 1A and 3A. I was curious so I looked at the brackets for all 32 sectionals. There were only 4 sectionals (Carroll, Zionsville, Hamilton Heights, and Northview) where every weight class had at least 8 wrestlers entered. There were also 4 that had at least 10 weight classes with less than 8 wrestlers. In total there were 160 weight classes over the 32 sectionals that 7 or fewer wrestlers, an average of 5 per sectional. The number of small weights doesn't necessarily relate to the multiplier, either. Two of the sectionals with complete brackets in all weights have higher factors, but one, Hamilton Heights, is considered :average" with a .97 factor. And one of the sectionals with more than 10 small brackets, Mishawaka, had a 1.18 factor. All of that means that at least one of the wrestlers in those weight classes reached the 5th/6th place match without defeating an actual opponent. Their only win was a forfeit. I didn't look at the individual matches to see how these wrestlers were distributed across the class sizes, but I'd guess more of them were class 1A wrestlers tn 2A or 3A. Since so many wrestlers didn't really earn a 5/6 place it seems like the value should be less, like half-way between a sectional entry and a regional entry. Link to comment Share on other sites More sharing options...
decbell1 Posted February 13, 2014 Share Posted February 13, 2014 You should have to beat an actual opponent before you move up from the baseline score Link to comment Share on other sites More sharing options...
maligned Posted February 13, 2014 Author Share Posted February 13, 2014 I try to post every one of AC's results if someone else hasn't already done so, but I see very few results for a lot of other class A teams. How does that affect the Genius Ratings? If a team only reports its wins will it help their rating? I also think that wrestling does not lend itself to this type pf prediction model as basketball or football. In those sports a coach can change his line up, but the team still plays together as a unit to achieve victory. In wrestling its 14 different individual matches that determine the result and a line-up change, like substituting a JV wrestlers to take a forfeit and moving your wrestler up against a very good wrestler from another team can have a greater effect than a line change in the other sports. I think that's why in wrestling more than any other sport you can't compare the team results against common opponents and predict results. You could report only your wins and still not impact your rating because teams tend to be relatively consistent. Your winning margins tell us just as much about your possible weaknesses as your losing margins. If I lose by 20 to the #70 team, it's a better performance than a 50-point win against #300. And we had 80% of all the results, meaning most of your results will be found out, even if you're not the one reporting them. In baseball, a team's quality varies greatly from game to game based on the pitchers. That doesn't mean we can't determine the average quality of a team over time. it also doesn't mean we can't design a model that shows the individual impact of each player on a team's success. We absolutely can do it. It's the same in wrestling. We can begin to define a team's average quality over time very effectively after only 5-10 matches, despite the fact that results can vary some with specific matchups. And common scores between teams are very predictive of future results. Of course you might isolate a specific example where A beat B who beat C who beat A, but over time, we can come to very accurate conclusions about your average performance level. And if we can do that, and we can simultaneously consider all the components of all the teams in the state that comprise those average performance levels, we can absolutely begin to define the impact of having certain quality levels of individuals on your team. Wrestling has its quirks, as any sport does, but it's very quantifiable when you have enough information. I definitely understand the questions with the multipliers and the desire to see the system in 1A work as well as the other classes. And methods for determining and adjusting for strength in teams' tournament path is something I'll always agree should be analyzed and critiqued. I'm just trying to point out that using Genius as a key component in getting information we need is very reliable and is not the source of our troubles (I can say this with confidence because I just administer a method and study its accuracy...I'm not the originator of its principles.) Link to comment Share on other sites More sharing options...
maligned Posted February 13, 2014 Author Share Posted February 13, 2014 You should have to beat an actual opponent before you move up from the baseline score This sounds right in theory, but when you run all the data--especially for 1A and 2A--there's a very mathematically significant positive impact on a team's performance level next year with relation to having someone in the lineup (even a non-placer at sectional) versus having a forfeit. And there's a very significant level difference between having a 5/6 placer vs. a non-placer. Of course there will be weights at every sectional where a 5/6 finisher did little to nothing to get his spot or a non-placer will seemingly never win a wrestling match that wasn't a forfeit. But the averages and their significance--especially at 1A and 2A--say that there are big jumps in predictive expectation for next year when you've got those types of guys in the lineup rather than nothing. Link to comment Share on other sites More sharing options...
decbell1 Posted February 14, 2014 Share Posted February 14, 2014 This sounds right in theory, but when you run all the data--especially for 1A and 2A--there's a very mathematically significant positive impact on a team's performance level next year with relation to having someone in the lineup (even a non-placer at sectional) versus having a forfeit. And there's a very significant level difference between having a 5/6 placer vs. a non-placer. Of course there will be weights at every sectional where a 5/6 finisher did little to nothing to get his spot or a non-placer will seemingly never win a wrestling match that wasn't a forfeit. But the averages and their significance--especially at 1A and 2A--say that there are big jumps in predictive expectation for next year when you've got those types of guys in the lineup rather than nothing. Right, but give "those types of guys" the Sec 5/6 points, but give only SECT points to guys if they finish 6th out of 6 or 5th out of 5 Link to comment Share on other sites More sharing options...
maligned Posted February 14, 2014 Author Share Posted February 14, 2014 Right, but give "those types of guys" the Sec 5/6 points, but give only SECT points to guys if they finish 6th out of 6 or 5th out of 5 Ok, I see what you mean...still get "warm body" points because they do mean something in 1A and 2A, but nothing more. I would want to collect data from a season on that type of wrestler and add them to the analysis to see how they project with the rest of the sec5/6 and sect-no-place results. Link to comment Share on other sites More sharing options...
AC Computer Posted February 14, 2014 Share Posted February 14, 2014 I have one other question about the multiplier. How far into the tournament does it project the individual's ability to advance? For instance, Jay County sectional has one of the lower multipliers and then in the regional they match up with Delta Which is slightly higher. Does the multiplier take into account that the chances to advance past the regional are slightly less because entries will be facing foes from a higher rated sectional? Or, vice versa, a wrestlers who advances from the Carroll sectional would face an opponent from a lower rated sectional, so his chances to advance would be slightly better. Would it be more accurate to have separate multipliers for each level? Link to comment Share on other sites More sharing options...
bog190 Posted February 14, 2014 Share Posted February 14, 2014 I have one other question about the multiplier. How far into the tournament does it project the individual's ability to advance? For instance, Jay County sectional has one of the lower multipliers and then in the regional they match up with Delta Which is slightly higher. Does the multiplier take into account that the chances to advance past the regional are slightly less because entries will be facing foes from a higher rated sectional? Or, vice versa, a wrestlers who advances from the Carroll sectional would face an opponent from a lower rated sectional, so his chances to advance would be slightly better. Would it be more accurate to have separate multipliers for each level? The multiplier considers the "entire path." Link to comment Share on other sites More sharing options...
JakeandBreighton Posted February 14, 2014 Share Posted February 14, 2014 Can you show a break down, by class size, for the 3 classes? Also, when classing them, for next year, do you use the out going senior class, or incoming freshman class? Link to comment Share on other sites More sharing options...
Recommended Posts