Froggy's High Correlation Picks

There isn't a bettor (non-degenerate) that anticipates OT. It's one of those rare occurrences that can affect the wager especially if you're on the UNDER. Therefore, when calculating results of the wager (for calculation purposes only), we have to consider just the 4 quarters only to maintain algorithm accuracy.

Sounds interesting.  I use a db as well and for totals use only regulation time as well.  Sent friend request.

Awesome start 

I remember you reaching out on one of my posts.. Seems like we have a similar way of doing things, though I can tell it's at least slightly different. Two heads are better than one.. Let me know if you ever want to communicate and work together. Like you said, results speak for themselves. I think it's clear that between my two threads and yours, we seem to be onto some good things.

@arakias, I agree that we're both capable of recognizing patterns when we see them. Martingale-type systems will eventually work, but the problem is the possibility of a longer-than-expected loss pattern. Even winning percentage of 75% can have 10 losses before a 30 win run, but by then the bankroll takes a hard hit. What I was thinking about when I saw your system is to take a few of the cappers here that are really solid and use your system to pick based on confluence (when more than one capper agrees on a game) ... this will probably have more consistent results than a blind pattern system. This is just my opinion, though.

Sounds like a winner.

I know @arakias may not want to take on this method because there's really no easy way to back test. The only way to do this is probably to pick long-term cappers and extract information from each of their posts. It's tedious, but if anyone is up to the task, you can probably get a couple of guys, break down the task and get the data that you need. For a winning system, I wouldn't shy away from the work.

If you broke it down by quarter/capper and you used 4 cappers

You would need 16 individuals to scrub the entire forum to get 1-years worth of data with each individual spending about an hour and half to attain this info. Then you can backtest the system quite easily for an entire year's worth of data.

Going forward, collecting this data each day would be minutes out of your day, with some profound results (obviously speculating)

@Froggy80

 I agree, that's why my base unit so low. I feel like the problem for 99% of people who use a LONG martingale is one of two things.

A) They don't have the bankroll for when the shit hits the fan.

B) They're not disciplined.

I've got both, so they work really well for me, particularly as you mentioned on systems that I can back test 10+ years of data. I'm like you.. minutes out of my day to research or comb through information is worth it for a winning system. 

I'm waiting for the winning lottery ticket to hit my hand... when (yes, I day dream a lot) that does, say bye bye day job and hello full time betting analytics.
I would love to do that now, but man, juggling things isn't my strong point and I get dis- SQUIRREL!! 
... Sorry, I mean to say I get distracted easily.
Also... I hear what you mean arakais... I don't have a bank roll and I'm def not disciplined... bad combo for me.
I've come into this betting world for less than a year now, and I've made those rookie mistakes way too many times... and I'm continuing to do so, but I do have to say I'm sloowwwlly getting better at it... very slowly though.

I do like how this whole forum is around with supportive people sharing their views and experience... and then of course, you got the dicks and aholes who just like to either troll or try to assert their dominance.
It's gambling - everything you do is your own responsibility - can't blame anyone, even if they say this is a guarantee and to bet your soul on this play.

Already have started looking at data to see if there is a way to fine tune this system.  Am willing to work with others.

I'm placing the over/under calculations on hold for a few days. I have a pretty good idea on how to proceed, but the calculations are a bit more complex than the ATS spread.

Frog are you back testing at all? How often do or would you have had multiple plays a day?

As the database grows over time (as I input more and more days from the past), the playable matchups increase. In the end, we should be able to predict every game available.

I am not posting plays that have less than 90% correlation to a past event. For example, yesterday the system picked MIA, but I didn't post it because the quality didn't meet my personal standards even if it was the right pick.

This data is not readily available. In fact, I only know of a single site that has the information that I use in order to produce these correlation calculations. So, I'm manually scraping this info. It's a lot of work -- but all things worthwhile are hard (that's what she... jk) 

I've uploaded 250 matches worth of data for the over/under and still correlation ratings are not near ideal. The highest I've seen thus far is 78% ... so I'll continue to upload more data until I get values of 90% or higher. In the meantime, I will focus only on ATS spreads for today.

Sometimes lives odds are better than prematch odds. No sweat if you miss a wager.

Any pick sir? Thanks 

GSW -5

DEN -9.5

DEN UNDER 226

O/U play received 88% correlation

Bol 

@Froggy80

 Do you have Twitter? Just suggesting since it would be easier to get picks with enough time to lock.

No bets for 6:30 matchups

you need help doing the manual stuff??or you dont wanna divulge whats involved in it?

may i ask how much data you are working off of?like how many seasons worth?...and do you feel that the rather unusual current circumstances may affect results??like just the different travelling schedules and no crowds watching etc?