One More Look at Regression Strategy

Setting and influencing the dice roll is just part of the picture. To beat the dice you have to know how to bet the dice. Whether you call it a "system," a "strategy," or just a way to play - this is the place to discuss it.

Moderators: 220Inside, DarthNater

Post Reply
User avatar
heavy
Site Admin
Posts: 10530
Joined: Mon Nov 22, 2010 11:46 am
Location: Texas
Contact:

One More Look at Regression Strategy

Post by heavy » Wed Dec 18, 2019 2:50 pm

One of the strategies many of us learned from the great John Patrick is the regression move. It’s a move particularly well-suited for DI’s with limited bankrolls. Simply put, a regression move is a play designed to lock up an early profit on a shooter and position the player for a larger win should the hand continue. It’s a strategy I use in virtually every session I play. While it does not give you a mathematical edge over the house, it does let the disciplined gambler play from a position of power. It is an especially effective way for precision shooters to play.

Let’s look at a player with a respectable SRR of 1:8 using the V-3 pre-set. For the sake of this example, let’s assume he keeps the dice on axis 100% of the time. Not only does the player know that on average, he’s going to get eight tosses in before the devil jumps up, he also knows the distribution of numbers that will likely roll. Out of sixteen possible on-axis combinations, two add up to seven, six add up to six or eight, four add up to five or nine, and four add up to three, four, ten and eleven. Our astute player knows his advantage is on the inside numbers, and he bets those numbers in proportion to his advantage, with three chips each on the six or eight for every two on the five or nine. He knows that he may go point-seven, or he may shoot the lights out with a forty-five number hand. But over the long haul, if his tracking is correct, he will average eight tosses once the point is established. To err on the conservative side, he’s elected to play a strategy that includes a regression to lock up a profit after the fourth post-Come Out toss. His regression involves coming down off the five and nine and reducing the size of his bets on the six and eight. He will “same bet” subsequent hits until he gets over his SRR “hump.” If the roll extends beyond roll eight he will press every other hit in an up and out strategy.

Now let’s plug some numbers in and see how the hand plays out. Let’s give the player $1000 bankroll in a $10 – 2X odds game. He plays $10 on the Pass Line and establishes the six as his point and takes $20 in free odds. He places the eight for $30, and the five and nine for $20 each. Remember, he sized his place action according to his advantage, placing more emphasis on the six and eight, where he has the greatest edge with the V-3. Now let’s play out the hand.

Six is the point.

Second toss – rolls 8. Shooter collects $35. $90 action on the table. $55 at risk.

Third toss – rolls 9. Shooter collects $28. $90 action on the table. $27 at risk.

Fourth toss – rolls 3. No change in bankroll. Shooter comes down off the 5 and 9. Regresses the eight to $12, and regresses the free odds behind the Pass Line to $10. $32 on the table. $31 guaranteed profit for the hand.

Fifth toss – rolls 10. Same action.

Sixth toss – rolls 8. Same bet. Lock up an additional $14. $32 action on the table. $45 guaranteed profit for the hand.

Seventh toss – rolls 5. Same action.

Eighth toss – 7 out. House locks up the $32 action. Player colors in a $45 profit.

It’s really that simple. Now, there are all sorts of variations you can spin off the above action. You could turn your bets off on roll’s seven, eight, and nine – leaving just your line bet working. You could pick up your free odds and just go with the flat bet for a few tosses. Or you could convert to Come betting and let the dice decide what do to next. Of course, if you choose the latter, remember you are taking chips out of your guarantee and putting them back in action. Limit yourself to one or two Come bets, or play a modified version of the Heat Seeking Craps Strategy by coming down on place action as Come bets travel, and going strictly with flat bets until a number repeats itself and “earns” its free odds bet.

Locking up a profit early is a smart way to play. If you will do it consistently, it will go a long way toward improving your craps bottom-line.
"Get in, get up, and get gone."
- Heavy

User avatar
Big O
Posts: 1457
Joined: Wed Apr 26, 2017 11:11 pm

Re: One More Look at Regression Strategy

Post by Big O » Wed Dec 18, 2019 9:31 pm

I have been working to improve my dont play as well as come up with a good regression play im comfortable with. So far i am progressing better on my lay betting than i am with my regression play. Putting up a lay bet when i feel the time is perfect has been more palatable than putting a sizable chunk of money at risk.
"if it was easy anyone could do it"

Moe Bettor
Posts: 1596
Joined: Mon Dec 21, 2015 2:31 pm

Re: One More Look at Regression Strategy

Post by Moe Bettor » Wed Dec 18, 2019 9:36 pm

Learned the regression play right here on this forum. One of the better lessons. Use it all the time.

memo
Posts: 870
Joined: Sat Jul 23, 2011 4:47 pm

Re: One More Look at Regression Strategy

Post by memo » Thu Dec 19, 2019 10:35 am

heavy wrote: Wed Dec 18, 2019 2:50 pm
Let’s look at a player with a respectable SRR of 1:8 using the V-3 pre-set. For the sake of this example, let’s assume he keeps the dice on axis 100% of the time.
Great example heavy. You know I base a great deal of my betting on Regression. This is a nice conservative approach that I may want to incorporate since my Bank roll has been depleted lately.

BTW...I want to be at the tables with anyone, anytime that is 100% on axis...Has to be a 'Perfect World Shooter'...
My thinking is that his SRR is way understated, or he has a huge double pitch problem.

Memo

Boxcars1212
Posts: 17
Joined: Sun Aug 07, 2011 6:44 pm

Re: One More Look at Regression Strategy

Post by Boxcars1212 » Mon Dec 30, 2019 12:19 pm

I use regression betting when I shoot the dice but not on others, and I never come down completely and will press after the regression 50% of my win on each hit. So if I bet $130 across in this scenario I would try for three hits and then regress to $52 across. I based this on my average rolls of box numbers but always qualify myself at the casino as it does take a few rolls to get dialed in to the table. I usually try and play a table where I can get the dice back quickly as it helps adjust to the table faster, I have a table at home that has a hard layout which mimics two of my local casinos. Durng my last casino trip my pressing strategy worked well as after my regression I threw five 5's and four 10's. Many times there seems to be a hot number or 2 so I like to press aggressively.

Clay
Posts: 117
Joined: Thu Feb 28, 2019 7:29 pm

Re: One More Look at Regression Strategy

Post by Clay » Wed Jan 15, 2020 8:38 pm

thnick wrote: Wed Dec 18, 2019 9:36 pm Learned the regression play right here on this forum. One of the better lessons. Use it all the time.
Agreed. So did I.

User avatar
Americraps
Posts: 1090
Joined: Mon Aug 08, 2011 2:42 pm
Location: Elgin, IL

Re: One More Look at Regression Strategy

Post by Americraps » Thu Jan 16, 2020 11:12 pm

My problem with regression is when you get stuck, it's harder to get unstuck.
See it in your mind FIRST...Then do it!

User avatar
Bankerdude80
Posts: 1895
Joined: Sat Jul 13, 2013 6:05 pm

Re: One More Look at Regression Strategy

Post by Bankerdude80 » Wed Jan 22, 2020 9:17 pm

heavy wrote: Wed Dec 18, 2019 2:50 pm Now let’s play out the hand.

Six is the point.
Second toss – rolls 8. Shooter collects $35. $90 action on the table. $55 at risk.
Third toss – rolls 9. Shooter collects $28. $90 action on the table. $27 at risk.
Fourth toss – rolls 3. No change in bankroll. Shooter comes down off the 5 and 9. Regresses the eight to $12, and regresses the free odds behind the Pass Line to $10. $32 on the table. $31 guaranteed profit for the hand.
Fifth toss – rolls 10. Same action.
Sixth toss – rolls 8. Same bet. Lock up an additional $14. $32 action on the table. $45 guaranteed profit for the hand.
Seventh toss – rolls 5. Same action.
Eighth toss – 7 out. House locks up the $32 action. Player colors in a $45 profit.
What does the regression bettor do when the first three numbers he tosses is non-paying trash or for that matter any combination of trash over the course of the eight rolls? Can this muck up your regression plans?
"Take the Money and Run...."
- Steve Miller Band

220Inside
Posts: 3406
Joined: Wed Nov 25, 2015 11:26 pm
Location: Massachusetts

Re: One More Look at Regression Strategy

Post by 220Inside » Wed Jan 22, 2020 9:33 pm

When I'm employing a regression strategy, I also use a threshold count where if the regression is not triggered by the count, I either do the regression or come down completely on the shooter depending on the tosses to that point.

User avatar
Big O
Posts: 1457
Joined: Wed Apr 26, 2017 11:11 pm

Re: One More Look at Regression Strategy

Post by Big O » Thu Jan 23, 2020 12:47 am

bankerdude80
What does the regression bettor do when the first three numbers he tosses is non-paying trash
i was working on a couple of regression strategies practicing today and ran into the same issue. I was wanting to do 3 hits and regress but i kept tossing horn numbers after i set the point. i did not come up with a suitable conclusion. I guess its like learning to play the dark side or lay numbers. You have to pay your dues.

I was just thinking today it has been a while since i noticed you posting.
"if it was easy anyone could do it"

Moe Bettor
Posts: 1596
Joined: Mon Dec 21, 2015 2:31 pm

Re: One More Look at Regression Strategy

Post by Moe Bettor » Thu Jan 23, 2020 8:43 am

IMO you just accept what is happening and move on. If that scenario of throws happens to me and I don't happen to be setting for horns, I don't bet except for the PL $10. The throw has to get right and the my numbers appear or screw it. And I have two betting styles. Early and late. Early is $12 6 and 8, one hit go to $18 each, next hit, take $21, back to $12 6 and 8. $13 at risk is acceptable for me with $34 action on the table. Haven't taken odds yet on my PL. Now we press to get to $50 for 1. And when we get to $50 for one we take that. Everything is paid for with profit. I will even regress from my $42 if trash starts coming.Late betting is $30 6 and 8, regress. Just a style I have that I feel protects the buy-in. Protecting the buy-in gives me freedom to go high. Don't know if I've explained this properly or it's just too conservative..but it has worked for me.

memo
Posts: 870
Joined: Sat Jul 23, 2011 4:47 pm

Re: One More Look at Regression Strategy

Post by memo » Thu Jan 23, 2020 9:51 am

"bankerdude80
What does the regression bettor do when the first three numbers he tosses is non-paying trash"

Note that in Heavy's example, there is one horn number....I have no problem seeing one..

My primary, go to set is the standard 3V..
If I start producing an inordinate number of horn numbers...(When I get on a streak, it can be more than 50% of the results)
I am double pitching or off axis as a group. It means that in the short run of variance, my on axis results is far below 44%, and I am solidly intrenched to the 7's side of the equation.
...Something is wrong, and I know it. You can bet that I am mentally trying to correct it, which means that I am tensed up, which means my muscle coordination is spiraling out of control. Tension and stress is not a positive thing for DI. I believe that I am not alone in this issue.

So, for the short run...
And yes, I understand volatility and variance...I understand validity of looking at the short term...I have also evaluated my shooting while practicing for thousands of hours.
I know that if I have four horn numbers with a few off axis sevens on the come out, I am more than likely going to have a short hand..This is a good time to re-asses the logic of regression at that moment and reconsider my bet...

I will also apply this logic to any other DI (unless he is employing GWAG)
Because....well, not because of superstition or any of that..
A DI who is attempting 7 avoidance with that result(s) is not on his game and has shifted to an area where 7 is more predominant. Our very practiced constancy will tend to keep us there. There is no better time to use Heavy's logic of waiting to see box numbers reappear before risking too much cash...

This is Kinda the opposite of see a horn, bet a horn. Which is also based on our tendency to do the same thing over and over.

Memo

DanF
Posts: 546
Joined: Sat Oct 01, 2016 4:33 pm

Re: One More Look at Regression Strategy

Post by DanF » Sat Feb 29, 2020 11:10 pm

My take would be not to use regression as a basic system, but a bonus on good runs.

1000$ being the bank I risk 230.

10$ pass, 20 odds. Independent of the box (30$)
Add 40$ 5&9, 60$ 6&8 (200$)

Two hits inside and down to 18$ 6&8, 10$ 9. Single odds pass. (36+20+10=66)
So you get 140 win or 112 plus a bonus from pass.
Best you get is 140+78...from two pass on eight say.
With a cost of 230, 218 win, 66 reinvest, 152$ locked after two hits being max pay, 112-66=44$ lock minimum.
Up a unit twice, powerpress, up a unit...

Something like that would be my plan now, for regression.
Gives me 4 shots possible in 1000$, I would chose 3 shooters I have good feel from, go with another plan for others...I think regression at the 30$ level is pointless. Not much reward for taken risk. Don't lock up enought after your move is done.
Last edited by DanF on Sun Mar 01, 2020 9:07 am, edited 2 times in total.

House of Orange
Posts: 1328
Joined: Wed Mar 07, 2012 11:09 am

Re: One More Look at Regression Strategy

Post by House of Orange » Sun Mar 01, 2020 1:37 am

It all works until it doesn't. Regression avoids depression.

Post Reply