A few -- eh, maybe more than a few, but not too many -- thoughts for anybody who is still playing.
Quote:
|
Originally Posted by roeldaboy
Of course its also important to be able to break your opponent, however other factors will influence a players skill nontheless, where service isn't and is just as good on hardcourt as on clay. On top of that I think the past has proven serve do be of major importance and I think you just underestimate its importance.
|
I wonder here if you are talking about real-life tennis on the ATP tour or tennis as represented in the game of Rocking Rackets? Obviously one is modeled on the other but they are different -- for example, real-life is hardly as predictable in terms of career development. Some of these differences are even arguably good in terms of making a more enjoyable game.
At any rate, I took a look at the documentation in the game when thinking about how to respond to this post and was surprised by a couple of things I found. It seems that information has been added that wasn't there before, as of several months ago when I came up with what has been called my '10 point player rating formula'. This is quite curious, since there literally hasn't been a post in the RR forums in a year and a half. It is now specifically defined for all to see how much home advantage, mentality, and surface preferences impact a match. This was quite surprising to me, and gave me an opportunity to lay aside some of my assumptions.
There are also some interesting statements which are not new about serving. It's stated to be only slightly less important than skill, and the trainer formula uses 75% of service with 100% of skill in the formula, both of which would indicate that my approach is flawed. With clear definition of the other elements, it seemed useful to me to conduct a test. It just so happens that my world presents a great scenario. The top three players over the past several years have played each other in the pro ranks 37-38 times. Actually just a bit more than that, but that's where it was when I took the match statistics are started to analyze it. The details are pretty complicated so I won't bother with a full writeup of them unless somebody is particularly interested, but I did come away with two primary conclusions.
**
Mentality isn't as important as I thought it was.. I had it pegged at 40% of the value of skill. According to the documentation it is only 30% when it is applied, which isn't most of the time. Key points like breakpoints and tiebreaks only. I thought I might be a bit high on mentality, but I didn't think I was off by that much.
**
I was undervaluing service, but not by much. Despite the documentation's statements about it's value, the actual description of what it does indicates it simply isn't close to skill in importance. It's far more important than any other rating, but not nearly as important as skill. Aside from not being used at all when the other player is serving(as obvious a thing as obvious gets), it is also weakened by the returner's speed. Against an elite player, who is going to have at least decent speed, that is not an insignificant fact, and there simply aren't enough aces to make it a close call.
New Player Rating Formula
So, having analyzed the statistics from dozens of matches making up the mentioned sample, I have come to the conclusion that I can somewhat improve my formula. I like having a 10-point scale but this is nearly a true 10-point scale -- in this case, nobody is going to exceed that. I'll be very impressed if anyone gets to 9.5 and 9.0 flat would be just fine for me. For comparison, Iglar ends up at 8.82 right now, and the sublime
Gaspare Cabrara from world 2 is, at slightly past his prime, recording a 9.29. He's a rare freak of nature, however. Here are the five elements of the new formula:
-- Skill * 0.95
-- Service * 0.515
-- Strength * 0.19
-- Mentality * 0.175
-- Speed * 0.17
I definitely understand how the game simulates matches a lot better now. As you can see here mentality is the big change, and as mentioned service is slightly more than half of the value of skill now. It's important to remember that some players overachieve with others underachieve. It is also true that service is slightly, but not all that much, more important for particularly old players since they don't have the speed to get as many returns in play and force as many rallies while receiving.
Hopefully this information will be of use to those of you still enjoying Rocking Rackets. I'll be making a few training adjustments, but not all that many -- thankfully it does not appear that I was very far off.