Does martingale really works?

[ranlo]

Initial BTC: 0.01
Start bet: 0.00000050

It takes 14 losses to lose, the chances of 14 losses in a row are small. The trick is NOT to expect doubling your initial BTC (quit at 0.015 or sooner, unless you're "feeling lucky"), AND not playing a pure Martingale. By this I mean that after you get let's say 11 losses in a row, and finally win, multiply your initial 0.00000050 by 10x-100x (or 1000x if you're "feeling lucky"), so you bet 0.00000500-0.00005000 for a few rounds (10+ losses in a row twice? Yeah, right... Just don't do too many bets with the new wager.), then go back to 0.00000050. This way it takes way less bets to get to your goal. Of course without a bot this is VERY boring and tiresome, so people start betting higher amounts, and therefore lose. Luckily there is https://bitcointalk.org/index.php?topic=307425

The bigger your initial BTC is, the bigger your start bet can be and the more mBTC you get faster instead of worthless satoshis.

The issue here is it still doesn't work. If you're going to risk it then you should just go with 50% chances on the full amount instead. Because it will be a faster way to either win or lose.

Posted From bitcointalk.org Android App

[1715245008]

1715245008

[1715245008]

1715245008

[1715245008]

1715245008

[1715245008]

1715245008

[1715245008]

1715245008

[C.Steven]

Initial BTC: 0.01
Start bet: 0.00000050

It takes 14 losses to lose, the chances of 14 losses in a row are small. The trick is NOT to expect doubling your initial BTC (quit at 0.015 or sooner, unless you're "feeling lucky"), AND not playing a pure Martingale. By this I mean that after you get let's say 11 losses in a row, and finally win, multiply your initial 0.00000050 by 10x-100x (or 1000x if you're "feeling lucky"), so you bet 0.00000500-0.00005000 for a few rounds (10+ losses in a row twice? Yeah, right... Just don't do too many bets with the new wager.), then go back to 0.00000050. This way it takes way less bets to get to your goal. Of course without a bot this is VERY boring and tiresome, so people start betting higher amounts, and therefore lose. Luckily there is https://bitcointalk.org/index.php?topic=307425

The bigger your initial BTC is, the bigger your start bet can be and the more mBTC you get faster instead of worthless satoshis.

For each win, you get 50 satoshi. So to reach 0.015 btc, you need to win 10000 times.

The chance of losing 14 times = 50.5% ^ 14 = 0.00007015833
The chance of winning 50 sat = 1 - 0.00007015833 = 0.99992984167
The chance of reaching 0.015 before busting =  0.99992984167 ^ 10000 = 0.49578747985

In contrast, you can all-in your 0.01 btc for just one single 1.5x bet with a win chance of 66%.

[ranlo]

Quote from: f3tus on Today at 05:24:26 AM

Initial BTC: 0.01
Start bet: 0.00000050

It takes 14 losses to lose, the chances of 14 losses in a row are small. The trick is NOT to expect doubling your initial BTC (quit at 0.015 or sooner, unless you're "feeling lucky"), AND not playing a pure Martingale. By this I mean that after you get let's say 11 losses in a row, and finally win, multiply your initial 0.00000050 by 10x-100x (or 1000x if you're "feeling lucky"), so you bet 0.00000500-0.00005000 for a few rounds (10+ losses in a row twice? Yeah, right... Just don't do too many bets with the new wager.), then go back to 0.00000050. This way it takes way less bets to get to your goal. Of course without a bot this is VERY boring and tiresome, so people start betting higher amounts, and therefore lose. Luckily there is https://bitcointalk.org/index.php?topic=307425

The bigger your initial BTC is, the bigger your start bet can be and the more mBTC you get faster instead of worthless satoshis.


For each win, you get 50 satoshi. So to reach 0.015 btc, you need to win 10000 times.

The chance of losing 14 times = 50.5% ^ 14 = 0.00007015833
The chance of winning 50 sat = 1 - 0.00007015833 = 0.99992984167
The chance of reaching 0.015 before busting =  0.99992984167 ^ 10000 = 0.49578747985

In contrast, you can all-in your 0.01 btc for just one single 1.5x bet with a win chance of 66%.

Or you can try a couple of 75% chances. The more repetitions you need the larger the chances of losing become. Not to mention the time that's involved.

Posted From bitcointalk.org Android App

[Cryptopher]

Martingale is a short-lived game, literally you can't use it for too long else the house edge catches up with you and loses you your bankroll.

It's more common than you think to see say 12 losses in a row in practice, than seeing 12 wins. Hell you rarely see five wins in a row.

[Phildo]

There is no way that you can do multiple -ev bets (any bet with a house edge is -ev) and make it +ev.

The problem with martingale is that even though you don't think you can lose x bets in a row (x is the number that will make you broke or hit the max bet), you can.

Then there's also the fact that it's stupid. I'll use whole numbers for ease.

bet 1, if you win you are up 1, if you lose you are down 1.
bet 2, if you win you are up 1, if you lose you are down 3.
bet 4, if you win you are up 1, if you lose you are down 7.
bet 8, if you win you are up 1, if you lose you are down 15.

Why risk 8, 16, 32, 64 to win 1 (on a 50/50 bet)?

[ranlo]

Martingale is a short-lived game, literally you can't use it for too long else the house edge catches up with you and loses you your bankroll.

It's more common than you think to see say 12 losses in a row in practice, than seeing 12 wins. Hell you rarely see five wins in a row.

If you're playing with 50% chances, you will see 12 wins just as often as 12 losses so I'm not seeing where that comes from. Even at 49.5% it's going to be pretty similar unless you're doing hundreds of millions of repetitions.

[C.Steven]

Then there's also the fact that it's stupid. I'll use whole numbers for ease.

bet 1, if you win you are up 1, if you lose you are down 1.
bet 2, if you win you are up 1, if you lose you are down 3.
bet 4, if you win you are up 1, if you lose you are down 7.
bet 8, if you win you are up 1, if you lose you are down 15.

Why risk 8, 16, 32, 64 to win 1 (on a 50/50 bet)?

The problem is that, many people wrongly believe a rare event can't happen.

Both a "x1.0101010101 bet with 99% winning" and a "x100 bet with 1% winning" have an expected value of zero, but I believe many people prefer the former one and they are probably thinking "how can I be that unlucky to lose in that?"

[f3tus]

If you're going to risk it then you should just go with 50% chances on the full amount instead. Because it will be a faster way to either win or lose.

Time doesn't matter if you're running a bot in the background while doing something else.

For each win, you get 50 satoshi. So to reach 0.015 btc, you need to win 10000 times.

No you don't, because you forgot me mentioning NOT playing a PURE Martingale. If you have 0.07 BTC instead of 0.01, you increase the losses needed from 14 to 16. I'm willing to risk 0.07 BTC for that tiny chance of getting 16 losses in a row.

[ranlo]

If you're going to risk it then you should just go with 50% chances on the full amount instead. Because it will be a faster way to either win or lose.

Time doesn't matter if you're running a bot in the background while doing something else.

For each win, you get 50 satoshi. So to reach 0.015 btc, you need to win 10000 times.

No you don't, because you forgot me mentioning NOT playing a PURE Martingale. If you have 0.04 BTC instead of 0.01, you increase the losses needed from 14 to 16. I'm willing to risk 0.04 BTC for that tiny chance of getting 16 losses in a row.

How much have you made using this strategy? I've run it ~12x total so far (starting at a single satoshi -- 0.00000001) and every time it has gone bust before doubling up. Note: didn't use Bitcoin though, I used Doge for testing. Same rates though.

[f3tus]

How much have you made using this strategy? I've run it ~12x total so far (starting at a single satoshi -- 0.00000001) and every time it has gone bust before doubling up. Note: didn't use Bitcoin though, I used Doge for testing. Same rates though.

I don't know because I was betting with insignificant amounts for testing purposes, maybe 0.01 profit altogether (not in a single day). I'm considering putting it to the real test when I get my next paycheck. And I'm sure you didn't play with my strategy - I play a modified Martingale as explained above.

[ranlo]

How much have you made using this strategy? I've run it ~12x total so far (starting at a single satoshi -- 0.00000001) and every time it has gone bust before doubling up. Note: didn't use Bitcoin though, I used Doge for testing. Same rates though.

I don't know because I was betting with insignificant amounts for testing purposes, maybe 0.01 profit altogether. I'm considering putting it to the real test when I get my next paycheck. And I'm sure you didn't play with my strategy - I play a modified Martingale as explained above.

No martingale is going to result in profit. There are a couple exploits that can be used on some sites to enhance the user odds, but even so it's still not resulting in doubling up before going bust, .

[f3tus]

No martingale is going to result in profit. There are a couple exploits that can be used on some sites to enhance the user odds, but even so it's still not resulting in doubling up before going bust, .

Profit != doubling up. You can still profit, it's just a question if you're happy with the result or not. I guess those with a gambler's mind wouldn't be satisfied until doubling or more, but I'm not that kind of gambler.

[ranlo]

No martingale is going to result in profit. There are a couple exploits that can be used on some sites to enhance the user odds, but even so it's still not resulting in doubling up before going bust, .

Profit != doubling up. You can still profit, it's just a question if you're happy with the result or not. I guess those with a gambler's mind wouldn't be satisfied until doubling or more, but I'm not that kind of gambler.

Then you could take 1 BTC, gamble it at 98.9%, win that 0.001 BTC or whatever and call that a successful betting strategy. It's only successful if it results, long-term, in earning more than you spent (which means you'd at LEAST double up).

[f3tus]

Then you could take 1 BTC, gamble it at 98.9%, win that 0.001 BTC or whatever and call that a successful betting strategy. It's only successful if it results, long-term, in earning more than you spent (which means you'd at LEAST double up).

I wouldn't risk 1 BTC to get that much when I can risk far less to get it. If I put 0.5 in and get 0.2 back and now have 0.7, I'd call that successful.

If you really want to gamble (daily) instead of play around (like me), feel free to do 7x bets and increase every x losses. Sure, it's more profitable, but you're also more likely to lose it than me.

[Cryptopher]

Martingale is a short-lived game, literally you can't use it for too long else the house edge catches up with you and loses you your bankroll.

It's more common than you think to see say 12 losses in a row in practice, than seeing 12 wins. Hell you rarely see five wins in a row.

If you're playing with 50% chances, you will see 12 wins just as often as 12 losses so I'm not seeing where that comes from. Even at 49.5% it's going to be pretty similar unless you're doing hundreds of millions of repetitions.

You would be surprised. On PrimeDice I've seen 12 consecutive losses so many times, but never seen more than 7 consecutive wins and that's with a 1% house edge.

But Martingale is definitely a loser in the long term, unless you're super patient maybe.

[C.Steven]

Then you could take 1 BTC, gamble it at 98.9%, win that 0.001 BTC or whatever and call that a successful betting strategy. It's only successful if it results, long-term, in earning more than you spent (which means you'd at LEAST double up).

I wouldn't risk 1 BTC to get that much when I can risk far less to get it. If I put 0.5 in and get 0.2 back and now have 0.7, I'd call that successful.

When there is a 1% house edge, you could expect a 1% loss on average for your wagered amount, no matter if you are doing 1% bet or 99% bet or martingale.

So the higher your wagered amount, the "worse" your strategy is.

[ranlo]

Martingale is a short-lived game, literally you can't use it for too long else the house edge catches up with you and loses you your bankroll.

It's more common than you think to see say 12 losses in a row in practice, than seeing 12 wins. Hell you rarely see five wins in a row.

If you're playing with 50% chances, you will see 12 wins just as often as 12 losses so I'm not seeing where that comes from. Even at 49.5% it's going to be pretty similar unless you're doing hundreds of millions of repetitions.

You would be surprised. On PrimeDice I've seen 12 consecutive losses so many times, but never seen more than 7 consecutive wins and that's with a 1% house edge.

But Martingale is definitely a loser in the long term, unless you're super patient maybe.

How often you see each short-term will depend on how the site handles them:

If it is 50/50 with 1.98x return, you should see both equally
If it is 49.5/50.5 with 2x return, you'll see one more than the other

You just happened to be extremely unlucky.

[C.Steven]

I personally have seen 15 consecutive losses on PD for x2 bets.

It is true that the probably of having 15 consecutive losses is very low, but if you run a bot for hours or days, the probability of having 15 consecutive losses in that tens of thousands of bets would not be small.

[Cryptopher]

Martingale is a short-lived game, literally you can't use it for too long else the house edge catches up with you and loses you your bankroll.

It's more common than you think to see say 12 losses in a row in practice, than seeing 12 wins. Hell you rarely see five wins in a row.

If you're playing with 50% chances, you will see 12 wins just as often as 12 losses so I'm not seeing where that comes from. Even at 49.5% it's going to be pretty similar unless you're doing hundreds of millions of repetitions.

You would be surprised. On PrimeDice I've seen 12 consecutive losses so many times, but never seen more than 7 consecutive wins and that's with a 1% house edge.

But Martingale is definitely a loser in the long term, unless you're super patient maybe.

How often you see each short-term will depend on how the site handles them:

If it is 50/50 with 1.98x return, you should see both equally
If it is 49.5/50.5 with 2x return, you'll see one more than the other

You just happened to be extremely unlucky.

That's the thing with Martingale though, all your losses could come in a row which is what rules you out from recovery. I think that Martingale works well when used intelligently, with discipline.

[f3tus]

It is true that the probably of having 15 consecutive losses is very low, but if you run a bot for hours or days, the probability of having 15 consecutive losses in that tens of thousands of bets would not be small.

That is true. But you do not play tens of thousands of bets with my modified Martingale where you switch the wager 10-100x for a few rounds here and there. It's riskier, true, but then again, it takes less bets, so maybe it's not. In either case, it's safer and better than a pure Martingale, which is a guaranteed way to fail.