Here's the answer to Wednesday's trivia question about the length of tenure for D1 women's soccer head coaches: What is the average tenure of today's D1 women's soccer head coaches? 7.15 years What is it for female head coaches? 6.39 years What is it for males? 7.5 years I'll have to come up with more questions. If you have questions that have data-driven answers, let us know, maybe I have the data to get the answers.
One thing that I've always wondered about is what the percentage is of players that stay on a roster for all four years or the entire length of eligibility. I doubt that there is available data that will answer that question. When my kids were looking at programs one of the things that we always did was to look at the rosters and try to determine what the retention rate was. If there was a high percentage of players that did not stay with it, it raised a red flag. We tried to find out what the reason was that there were a lot of departures - transfers, kids just quitting, or whatever other reasons...
From my daughters 4 years there has been typically 1-2 that fall out every year. The reasons for the drops range from non playing time to relationships to breaking team rules
Because when there was a longer time period people would go back and change posts, you know, to make them look better, for after the fact arguments. The powers that be decided that 30 minutes was long enough for someone to basically correct a typo. The time has been shorter. Otherwise, once you've said it, you're on the record.
One to two players per year seems to be a pretty low rate of attrition. I've heard of some programs that routinely lose more than that but, as I said, that was a red flag for us that required a bit more scrutiny into the causes.
Getting back to trivia questions, try this: If you are not a conference automatic qualifier for the NCAA Tournament, what is your percentage likelihood of getting an at large position if: 1. Your RPI rank is #35; 2. Your RPI rank is #45; and 3. Your RPI rank is #55?
Pretty darned good guesses by Soccerhunter. The following chart shows how the percentages come out, by ARPI rank, for teams that are not automatic qualifiers getting at large selections to the NCAA Tournament from 2007 to the present. The chart includes a polynomial order 2 trend line, which I think is a good representation of the data. It also includes the trend line formula (in the formula, for ARPI rank #30, x = 1, for #31, x = 2, etc. And, it includes an R squared value for the trend line, which is a measure of how well it fits the data. An R squared value of 1 would be perfect, and this particular R squared value indicates a pretty good fit. The table below the chart shows the percentage likelihood of an at large selection at each rank position, based on the trend line formula. One might take this a indicating the RPI is the most important factor the Committee considers in making at large selections. It no doubt is a very important factor. The way I look at it, however, is that the Committee is looking at teams’ profiles, which include quite a bit more than the RPI. If a team, apart from the RPI, has a strong profile, then it is going to end up near the top of the table. As teams’ profiles decline, they are going to move down on the table. Thus, at least to an extent, teams’ profiles (excluding the RPI) are going to be fairly close to matching where they fall in the RPI spectrum. In truth, the match isn’t perfect, but it’s pretty good.
Here is a question to muddle over, having to do with likelihood of getting an NCAA Tournament at large berth: Background: A conference’s teams finish the regular season in conference standing positions 1, 2, 3, and so on. For conferences with tournaments, the teams finish in positions 1, 2, 3.5 (semi-final losers, average of 3 and 4), 6.5 (quarter-final losers, average of 5 through 8) or 5.5 (average of 5 and 6 for conferences where there are only two quarter-final matches), and 9.5 (average of 9 and 10, for SEC and Southland which have play-in matches). One of the things I tally is the average of a team’s conference standing and conference tournament finishing positions, as a way to evaluate how it did in its conference. So, a team with an average conference standing of 1 was first in the regular season standings and won the conference tournament. A team with an average of 1.5 either was first in the regular season standings and runner-up in the conference tournament or vice versa. And so on. Question: For teams with RPI ranks in the #31 to #57 range (the at large bubble), that are not automatic qualifer conference tournament champions and are not disqualified due to winning percentages below 0.5000, what average conference standing is the most likely to get its team an at large selection? What is your reasoning?
The most likely Conference Regular Season and Tournament Combined Standing to get an at large selection, out of teams ranked #31 through #57 (and not an automatic qualifier or disqualifed team), has a combined standing of 4.75. When I first looked at this chart, it seemed not right. But on further thinking, it seems right. The kinds of questions I ask myself are, which teams are likely to be in the bubble and have a combined standing of 1.50, which are likely to have a combined standing of 4.75, and so on. Bubble teams with low combined standing numbers are likely to be from mid-majors. Bubble teams with combined standings in the 4.75 range are likely to be from Power 5 conferences.