Originally Posted by ohplease
Textbook example of why games aren't a good metric. There are LOTS of guys out there who've mastered tennis at their particular level, with no real interest in improving their games. And why should they? They get to pull a Shaquille O'neal and play possum during the regular season, taking a close loss here, scraping out a close win there, all in an effort to protect their *adj ratings.
Why *adj? Most likely because they won a lot at that same level last year, both locally and in the playoffs. Now they get to do it all over again this year. Same faces, over and over and over. Which is what you see in USTA play. You can practically see the thought bubbles over their heads: "please, please take this game. I'm trying to avoid strikes. Yeah, I know I hit a ridiculous diving backhand winner 10 minutes ago, but that frame shot into the next court? Nevermind that."
Now, by going to winning percentage, that all changes. Sure, people can manage percentages during the regular season, but if they want to make the playoffs, they need to bring it for real. And - if you make the playoffs - you're bumped.
This scenario would of course feature teams trying to out-tank each other to avoid the playoffs this year in the hopes of a favorable draw at their level next year, but there's only so much they can do, as there's going to be an influx of players from the next higher level being bumped down because they can't win there, but probably can here.
If you're a diehard 4.5 style player, but you can't win at 3.5, then you're demoted all the way down to 2.5 if need be, until you figure out how to win. If you're the unbeatable amateur grinder at 3.0/3.5, you get promoted to 4.0/4.5 and forced to learn some new tricks.
In my own experience, I'm seeing guys with winning percentages above 80%, on teams that finished in the top 4 of around 20, who aren't getting ESR bumps. Why? Probably because they get killed by teams built for sectional/districts/etc., who are beating guys otherwise dominating their level 6-1, 6-2.
This is all well and good if you believe that someone cant possibly win 80% or more of their matches and still be within the correct rating.
I think changing the system just because some people happen to be sandbaggers doesnt make a lot of sense. Sandbaggers aside the system works fine for most people in the league.
If the league wants to do something about these people who are only there to cheat the system, they should send some people to the matches or do something specifically to target them, rather than ruin it for everyone else in the league.
A good majority of players dont even play enough matches every year (here the average is like 5-6) to even be considered under any sort of 80% win rule, so you have to go by games to get more granular.
And besides the games factor your precious 80% rule doesnt factor in where people are within their own rating. If they just happen to play weaker 3.5's should they be punished for that? It doesnt make any sense to me.
I think this is just a simplistic approach that doesnt consider what the skill ratings really mean or what winning a tennis match is all about for that matter. (for most people, not counting the sandbaggers)
Im sure this isnt perfect either, but I think a better sceniro would be to "reward" the teams that move on to districts by using some form of the "move up / split up" rule on them like if you finish in the top four in nationals. Because think about it, why do players want to sandbag?
Some of them are sick and twisted and only want to win matches no matter who they play, but others simply want to win the league which is fair (it is a league after all and your goal is to win it.....). So maybe if your team moves on, more people should get rated up than if not (versus how it seems to work now where you are less likely to get rated up because you got spanked by the bigger sandbaggers from some other district/section).
Then maybe some win percentage rule is in order, or some variation of it. But to just institute a win percentage for everyone is not a good idea in my opinion.
If you think people should be zipping up and down from 4.5 to 2.5 from year to year, you really dont have a realistic idea of what skill ratings really mean.