I would agree with you Ratty if you were talking about a WR 500 player beating a WR 50 player.

That is because those are rankings with a lot of points, thus (almost certainly) gained over a range of tournaments. As such, they will be a 'true' reflection of that player's level over a period of time.

But any player ranked over 1000, with barely a few points (Harry had FIVE points), is another story. That can be from one or two tournaments; in itself, it tells you nothing meaningful about their level. (Obviously, if they've played 25 tournaments and got 5 points, that's different).

As such, RJA has a point - a youngster who has only played a couple of tournaments, and got a couple of points, does not have a ranking that is relevant in any real way.

Of course, Harry may lose tomorrow. Of course, it's wrong to say he is really WR 300 level because he beat one today. (and he's played about 8 tournaments, not 2). But he is not necessarily WR 1200 either. If Andy played no tennis for a year and was unranked, he wouldn't be an unranked level player and wouldn't regress to the mean of unrankedness after his first good win.

This obviously applies to many youngsters who zoom up the rankings. It also applies to US college players when they start - their ranking based on a couple of summer tournaments is of very little value in predicting their real ranking level.

Ratty wrote:

---

RJA wrote:

---

Ratty wrote:

---

paulisi wrote:

---

Harry could be a case of after the lord mayor's show. Too many times have players failed to follow up after massive upsets.

---

Yep, no matter what our much-respected co-poster Indiana might say, it's called REGRESSION TO THE MEAN ...

---

I am genuinely curious as to how you think that a guy that has only just turned 19 could have established a meaningful mean.

---

If a guy ranked 1242 beats a guy ranked 283, there are surely only 2-ish explanations: either that 1242 played above his standard, or that 283 played below his standard. Or - more likely - a mixture of both. 

The alternative explanation is that 1242 has improved his standard significantly in a very short period of time. To me that doesn't seem very likely.

Hence, if one accepts my hypothesis that Harry played above his standard, then it's pretty likely that he will revert back again fairly soon - maybe tomorrow, maybe next week. Or maybe, like Dan Evans last year, he'll keep it up for 4 weeks. And when he does, then just like with Dan Evans, there'll be people coming up with complicated explanations as to why he hasn't "kicked on" or whatever.

And I really fail to see why this is controversial. Yeah, I'm an ordinary Joe with no particular expertise in human behaviours. But I read books by people who do have this expertise; and "outperformance" followed by "regression to the mean" in sports players is just mainstream stuff.

(And of course, it only bears out paulisi's anecdotal observation.)

 

---

 So what, players never improve?

Ratty wrote:

---

Er, I do agree with the concept of what you are saying CD, but how can I put this politely ... Harry has played 22 tournaments in the last year. 

---

And, I really don't see what that has to do with anything given that he is 19. At that age it is not uncommon for players to make sudden and dramatic improvements. I don't know a great deal about Harry but I have seen him a few times over the last couple of years. Even at 17 he had a very big serve albeit it an inconsistent one, possibly in part due to an unorthodox service motion. I can easily understand how a few relatively small improvements to his game would lead to considerably better results, especially on indoor courts where a big serve is more valuable asset.

indiana wrote:

---

Yes, another Rattyism.

---

Believe me, I know that nothing is more pointless and tedious than trying to "win" an argument on a Message Board.

But to repeat, what I said is mainstream stuff. And let's not call it a Rattyism. Let's call it a "Kahnemanism" - after, you know, the legendary Nobel-Prize-Winning behavioural psychologist. You know, a chap who writes books that we pseudo-intellectuals read (actually his last one topped the charts):

Regression effects are ubiquitous, and so are misguided causal effects to explain them. A well-known example is the "Sports Illustrated jinx", the claim that an athlete whose picture appears on the cover of the magazine is doomed to perform poorly the following season.

Ratty wrote:

---

indiana wrote:

---

Yes, another Rattyism.

---

Believe me, I know that nothing is more pointless and tedious than trying to "win" an argument on a Message Board.

But to repeat, what I said is mainstream stuff. And let's not call it a Rattyism. Let's call it a "Kahnemanism" - after, you know, the legendary Nobel-Prize-Winning behavioural psychologist. You know, a chap who writes books that we pseudo-intellectuals read (actually his last one topped the charts):

Regression effects are ubiquitous, and so are misguided causal effects to explain them. A well-known example is the "Sports Illustrated jinx", the claim that an athlete whose picture appears on the cover of the magazine is doomed to perform poorly the following season.

---

You seem determined to miss the point. Nobody is arguing against the idea of regression effects, merely saying that they don't apply to each and every case you choose to apply them to. A developing young tennis player does not really have a mean level of performance that he may rise above and then regress back to. I suspect that Daniel Kaheman understands this, you evidently do not.

Ed C was down a break early in the set at 0-2* but has fought back to 2-2*

Bob in Spain wrote:

---

The battle of the Dans was supposed to start at the same time as Ed but no scores have appeared yet. Let' hope it is a technical malfunction or a delay and not a retirement.

---

The latter, I'm afraid, Bob:

QF:  (5) Dan Smethurst WR 377 "defeated" (1) Dan Cox WR 249 by a walkover

Hope it's nowt serious.