How to Handicap College Basketball by 70kgman

[70kgman]

1. Isolate only the home teams between -5 and +5. check!

2. After step 1, find the home teams that have played a equal or larger differential of games on the road than at home. i.e Team A (home team) is 10-6 overall with 6 of those 16 being home games, team B is 12-4 overall with 8 of those 16 being home games. Team A has the +2 differential. Check! Move on.

3. Check strength of schedule for both teams. I personally use Saragin ratings SOS for this rather than the official strength of schedule. If the home teams that made it through steps 1 & 2 have had the more difficult schedule, check! They move on.

4. Now moving on to stats. Check FG%, FG% Against, Average rebounding margin per game, and 3-point%. If the Home team is better than the opponent in all four of those stat categories, they move on to the final step.

5. Injuries. As long as the teams have a clean injury report, or at least a mostly insignificant one. They are good to go.

I wait until December before I start playing wagering to lets some stats accumulate and balance out a bit first. I personally play the favorites on the ML to win 1 unit, and for the dogs, I play 1 unit on the spread, and a 1/2 unit on the ML. I use those steps and stat categories without so many rules to cap games in general, but the results for the games that have fit the criteria exactly as I described have been amazing this season. You are not going to get a ton of plays using this capping strategy, maybe 30 or so a season, but worth the effort when you see the results.

[aajreyno]

What kind of results had this method produced?

[phans3]

You are not going to get a ton of plays using this capping strategy, maybe 30 or so a season, but worth the effort when you see the results.

the point is that it will be hard to know if your system is ev+ in the longrun with such a small samplesize

did you any backtesting over a decent samplesize?

[70kgman]

getting grilled already? It's not like I posted some whacked out system that pulls games seemingly out of nowhere. It is just using the most relevant stats in capping games (imo) to find some bad lines. You will see it ends up mostly applying to the Samford vs. Furman type games that involves relatively unknown teams where the linesmakers are probably more likely to be off. This is not something that can be back-tested since it deals with real time stats. I have used this capping method for a coule years now with great results. Haven't documented my record anywhere so I don't have any exact record to claim. I would guesstimate it hits around 75% of the time. Use it or don't. Just thought I'd share.

[Peregrine Stoop]

datamining

[daprospecta]

Cool

[jgilmartin]

datamining

This.

[bztips]

datamining

Yep.
Applies to just about all of the "systems" offered up by people who think they've found a magic bullet.

[Maverick22]

Yep.
Applies to just about all of the "systems" offered up by people who think they've found a magic bullet.

All Data Mining Isnt Bad...

[dialup_king]

the point is that it will be hard to know if your system is ev+ in the longrun with such a small samplesize

did you any backtesting over a decent samplesize?

HE can look at cloing line value also. College bball is softer than other markets

[Richards]

Thanks for posting 70kgman. Even if one doesn't buy into or disagrees with the effectiveness of your method, it's always nice to see what someone else is doing. Hopefully you'll get some good feedback.

[wesleysnipes]

that's a fair strategy. but i think its only 10 percent good. you left out 90 percent of the other stuff to do... its way more to handicapping college basketball than that. what else? you have to look at matchup edges on covers.com on all the games, you have to look at trends on all the games, you have to look at the charts on SBRforum.com- and read all the info on the games- from the offense/defense/charts/head to head matchups/away/home stats,etc...../ you have to read articles on the games/smart money from the cappers on all the forum of therxforum.com,covers.com,sbrforum.com, you have to fail for a longggg time before you get the hang of it. you have to jus know how to bet. like staying away from totals, like not betting high favorites, like not betting min small favorites, like not betting underdogs, like money management, like betting 1 game/day vs 10 games/day, like knowing how to use the information the best way to pick the best game. it takes time. so you have a small portion of how to handicap sports. im 20- and i still have a lot to learn. but i would say im pretty intelligent on handicapping. lol.

[rfr3sh]

that's a fair strategy. but i think its only 10 percent good. you left out 90 percent of the other stuff to do... its way more to handicapping college basketball than that. what else? you have to look at matchup edges on covers.com on all the games, you have to look at trends on all the games, you have to look at the charts on SBRforum.com- and read all the info on the games- from the offense/defense/charts/head to head matchups/away/home stats,etc...../ you have to read articles on the games/smart money from the cappers on all the forum of therxforum.com,covers.com,sbrforum.com, you have to fail for a longggg time before you get the hang of it. you have to jus know how to bet. like staying away from totals, like not betting high favorites, like not betting min small favorites, like not betting underdogs, like money management, like betting 1 game/day vs 10 games/day, like knowing how to use the information the best way to pick the best game. it takes time. so you have a small portion of how to handicap sports. im 20- and i still have a lot to learn. but i would say im pretty intelligent on handicapping. lol.

[That Foreign Guy]

All Data Mining Isnt Bad...

IF every part of the data mined criteria make sense then you have something that is worth testing against fresh data.

Step One, fine it might make sense that games the market expects to be close might produce more +EV opportunities (there are convincing arguments in both directions so some might be overweighted resulting in value on the other team).

Step Two, does not compute. Why does it matter that the home team has played less games at home than the away team? This seems arbitrary even ignoring the fact that Home court advantage is smaller in basketball than any other major sport.

[70kgman]

Hmm, I posted this last season. Why did it get resurfaced? Anyway, I still use basically the same strategy to cap ncaab games, I am just not so black and white about the steps anymore. SOS and amount of games played home/away are the most important angles of it in my opinion.

Step Two, does not compute. Why does it matter that the home team has played less games at home than the away team? This seems arbitrary even ignoring the fact that Home court advantage is smaller in basketball than any other major sport.

As far as this comment, that capping method involves statistics. Teams play better at home, if the opponent has played 7 of their 10 games at home, their stats are going to be slightly skewed and look a little better than they really are. And vice versa.

If a team that has played 7 of their 10 games at home is playing is playing a team that has played 7 of their 10 games on the road, they have a similar SOS, similar stats in every category, and the line is a pick. I am taking the team that has put up those same numbers mainly on the road every time. Those are the kind of lines I look for.

p.s The title of this thread was a parody of another thread in the think tank at the time I posted this a year ago. No longer relevant, so if it comes across at odd...

[That Foreign Guy]

That makes sense, I wonder if a venue-weighted SOS rating is a better way to approach this though?

[mental]

I don't see anything wrong with data mining. Stats do show trends and some people like trend betting (not me). Yikes this is an old thread ;p

[chunk]

All Data Mining Isnt Bad...

This.

[Thunders77]

Looks like however simple your approach is, It is producing Very Good Results!!

Best Regards

[Sawyer]

How its going fellas? Can u post plays? Whats the record ytd?

[70kgman]

Hmm, I posted this years ago. Was a stat "system" I tested/used for a couple years that did well. Maybe it is still relevant, but I am no longer using it personally. This thread should probably just die.