[SBR Lou]

-MUST get good numbers
-MUST have database of info
-MUST have programmable model
-MUST have hundreds of available filters
-MUST continually scrape data from numerous information portals

[Shark79]

And you do all of the above ?

[onthewhat]

Fading LazyLou

[mathdotcom]

You forgot one:

-MUST win at least once in a while


How are you liking your -318 on Federer by the way?

[Karayilan9]

There is an art to the science Lou, the data only goes so far, also make sure you don't ignore the psychology involved in the game which is often overlooked. As its a 1vs1 sport the psychology of players is very important as there is nobody to make up for their emotional instability unlike a team sport.

[EaglesPhan36]

I think you're looking at too many stats Lou. There are several that are important, court surface/speed, including serve %, points won on 1st/2nd serves, return % and break points converted. Of course, records at the court being played on and current form. Once you get past those, you're really getting into the nitty gritty and IMO drowning in too many minute stats.

[reno cool]

I've never seen anyone grow so much confidence with each loss.

[EaglesPhan36]

I think an underrated part of capping any sport is actually watching the games/matches too. Sometimes players win and advance in tennis matches and may have been outplayed despite what the stats might tend to indicate.

[englishmike]

Win or lose, I take my hat off to Lou's Resilience.

The ridicule of recent days would have crushed a lesser man, yet the guy is back in the game, head down, marching along the gambling highway.

He's due a good run, I'm rooting for winners for Lou.

[Matt Rain]

CrazyLou:

-GUESSTIMATES what a good number is
-LOOKS quickly at the ATP/WTA rankings
-SIMULATES matches on his Xbox 360
-FILTERS out ugly/unkown players
-SCRAPES a living, but not with his tennis bets

(I kiiiid.)

[tacomax]

-GUESSTIMATES what a good number is
-LOOKS quickly at the ATP/WTA rankings
-SIMULATES matches on his Xbox 360
-FILTERS out ugly/unkown players
-SCRAPES a living, but not with his tennis bets

How close is that to your model, Louie?

[pavyracer]

I think if you fade all his picks with players last names ending in "c", "a" & "i" you can retire in one year. This is the model pavyracer is using.

[betplom]

I think if you fade all his picks with players last names ending in "c", "a" & "i" you can retire in one year. This is the model pavyracer is using.

[daggerkobe]

I've never seen anyone grow so much confidence with each loss.

He's employing Brandon Lang's Jedi Mind Trick........ no record = no losses. Treat everyday like you're on a 20-2 run even tho the reality is 2-20.

[yisman]

CrazyLou:

-GUESSTIMATES what a good number is
-LOOKS quickly at the ATP/WTA rankings
-SIMULATES matches on his Xbox 360
-FILTERS out ugly/unkown players
-SCRAPES a living, but not with his tennis bets

This contains a lot of truth.

He's employing Brandon Lang's Jedi Mind Trick........ no record = no losses. Treat everyday like you're on a 20-2 run even tho the reality is 2-20.

Or keep claiming it's a small sample size when you lose. When you win, say it's because you're a brilliant handicapper.

The sample size will never be big enough for Lou to admit he's clueless about handicapping tennis.

[TurkzZ]

I think an underrated part of capping any sport is actually watching the games/matches too. Sometimes players win and advance in tennis matches and may have been outplayed despite what the stats might tend to indicate.

Hit the nail on the head there Eagels

[twister]

I think an underrated part of capping any sport is actually watching the games/matches too. Sometimes players win and advance in tennis matches and may have been outplayed despite what the stats might tend to indicate.

That's why including "total points won" is very important in a sport like tennis.

I've seen matches where one guy has won the majority of his service games to 15, whilst taking the opponent to deuce more often but not converting enough break-points. Then the opponent breaks on first/second opportunity more often than not and gets the win.

I've noticed that players that have won whilst scoring less points than their opponent in their previous match/es, that go on to face a player who scored more than his opponent in his previous match/es, will lose more often than not.

Whilst you can't use this information solely on its own, it is one way to determine "form" more in-depth than just looking at the results.

[jjgold]

If you played high level tennis and traveled circuits around world you might have a shot

Or...... Know qualifiers on tour

[Dark Horse]

-MUST have hundreds of available filters

There's your problem.

[SSLP]

lol this thread made laugh

Lou Lang

[SSLP]

I've never seen anyone grow so much confidence with each loss.

amazing post , strike of genius

[Justin7]

After 400 plays: you've beaten the market consistently, and you're winning.

[Dark Horse]

My impression is that CL has created some sort of model, has data-mined it to the max, and has convinced himself that his inflated numbers prove him to be a great capper; regardless of the results! This would explain why he sees these huge advantage plays, as well as why all these losses in no way affect his conviction.

[SBR Lou]

My impression is that CL has created some sort of model, has data-mined it to the max, and has convinced himself that his inflated numbers prove him to be a great capper; regardless of the results! This would explain why he sees these huge advantage plays, as well as why all these losses in no way affect his conviction.

Not exactly. Just because you have the filters doesn't mean all are always applicable, there are different things to look for in each situation.

[Justin7]

Not exactly. Just because you have the filters doesn't mean all are always applicable, there are different things to look for in each situation.

Filters and adjustments should be universally applicable. When you use them sometimes and not others, you introduce data mining errors.

[SBR Lou]

Filters and adjustments should be universally applicable. When you use them sometimes and not others, you introduce data mining errors.

Are you suggesting I should pay attention to how a player performs on grass when the event is on clay, or against a left-handed player when up against a righty?

[Dark Horse]

You mentioned hundreds of filters. That sounds a little different than standard categories such as playing surfaces, and suggests that you use filters to fit your purpose (that purpose being to create a more or less ideal model).

[SBR Lou]

You mentioned hundreds of filters. That sounds a little different than standard categories such as playing surfaces, and suggests that you use filters to fit your purpose (that purpose being to create a more or less ideal model).

Those are one of many filters.

[Justin7]

Are you suggesting I should pay attention to how a player performs on grass when the event is on clay, or against a left-handed player when up against a righty?

I'd suggest that grass and clay are different sports. Not just a filter.

[Wheell]

There is some are to analyzing tennis results, and weighing the importance of results on other surfaces is tricky. The correct weight isn't 1 (identical surface) and it isn't 0 (completely irrelevant). I must say weighing off surface results is probably one of the weaker areas of my tennis handicapping, but I don't know if I weigh them too heavily or not heavily enough.

[HeeeHAWWWW]

I'd suggest that grass and clay are different sports. Not just a filter.

They're different sports for Davydenko and other grass muppets. Same sport for Federer though :-)

[Justin7]

They're different sports for Davydenko and other grass muppets. Same sport for Federer though :-)

How does he stack up on clay versus grass?

I think Federer is overvalued. He dominates other surfaces, but the clay specialists have value. I have no stats to back up this opinion, however

[SBR Lou]

I think Federer is overvalued. He dominates other surfaces, but the clay specialists have value.

He's not usually talked up for his clay-court skills, but he's 28-0 since 2005 at Roland Garros against everyone except Rafael Nadal. He made the semis in 2005 and the last three years he lost in the finals, and is in the semis this year.

[Justin7]

3 years ago, I did a lot of tennis betting. I worked with that Russian who was featured in WSJ (and dumbly admitted that he tried to fix a few matches). We used a stat-based approach, and it worked very well.

Grass and hardcourt are somewhat similar. Clay is a completely different sport. But don't take my word for it - do the statistical regressions yourself. Look at the error ratings when you assume court-specific stats on other surfaces. Going to and from Clay is the worst (and grass to clay is the worst of the worst).