I’ve recently started re-reading an old Frank Scoblete book called Guerrilla Gambling: How to Beat the Casinos at Their Own Games. It’s full of information, some good, some lousy. Scoblete is an entertaining writer, and much of what he writes is accurate.
Even the inaccurate stuff he writes has the virtue of being entertaining. One of my favorite parts of the book, though, is the section where he describes his bankroll management strategy for a “guerrilla attack” on the casino.
In this post, I’d like to take a look at his recommendations to see how practical and accurate they are.
Scared Money Always Loses
His first bit of advice is that scared money always loses. You need to have enough money in your bankroll for your trip, and you also need to know when to quit.
It’s hard to argue with any of this advice. If you’re playing with scared money, you’ll often make bad decisions that cost you money in the long run. That’s about as practical as it gets.
My advice about a gambling bankroll starts with this:
This includes money you need to pay the rent with, money you need to pay a credit card bill, or money you need for a child support payment.
If you’ve ever seen the movie Indecent Proposal, you probably already know that the casino isn’t in the business of bailing out people from their financial problems. (Interesting film, by the way. Woody Harrelson and Demi Moore go to Vegas to try to win enough money to save a piece of real estate. Spoiler alert—they lose. Then Robert Redford offers them a million dollars for a night with Demi Moore.)
Scoblete also mentions that this kind of advice is vague, so he offers to provide some more specific and practical tips for how to manage your gambling bankroll. This advice includes which games you should play and how much money you’ll need for each of them.
Here’s Scoblete’s Bankroll Advice for Blackjack
Single Deck Blackjack Bankroll Requirements for Card Counters
If you’re counting cards in a single deck game, Scoblete suggests that you start with a betting spread of 1 to 6 units. He also suggest that you don’t start off betting a single unit; instead, you start off betting 3 units and adjust that amount according to the true count.
If the true count is +1, you bet 4 units, and if the true count is +2, you bet 6 units. If the count is -1, you bet 2 units, and if the count is -2, you bet 1 unit.
He goes on to suggest that you have between 100 and 120 units to play with in a single session, but you need 1500 units in your total blackjack bankroll.
What does this mean when you convert it into dollar amounts?
Let’s say you’re comfortable with a betting unit of $10. You’d start by betting $30 per hand. You’d bet between $10 and $60 per hand, based on the count. You’d also need a session bankroll of between $1000 and $1200, but your total blackjack bankroll should be $15,000.
From what I’ve read about blackjack bankroll requirements for card counters, these numbers are off in various ways.
The idea is that by having mini-bankrolls, you can improve your odds of winning by quitting when you’ve lost a certain percentage or when you’ve won a certain percentage. (There’s not much reason, otherwise, to have a “session bankroll.”)
But having separate session bankrolls is unnecessary and unhelpful. It lends credence to the belief that the odds somehow change from session to session. But mathematically speaking, we’re all playing one lifelong gambling session, and that session will eventually reflect the mathematical expectation.
The suggestion that you start with a 3-unit bet is also questionable. All the card counting books I’ve read suggest betting a single unit to begin with and only increasing that amount when the count becomes positive.
His book doesn’t explain why you would start with a 3-unit bet and only drop to a 1-unit bet when the count is negative. I suspect that it’s got something to do with camouflaging your, but I’m skeptical as to whether this technique would fool anyone working for a casino.
Finally, the overall bankroll requirement of 1500 units is excessive. If you’re a reasonably competent card counter, a bankroll of 1000 units offers a risk of ruin of only 1%. I don’t think you need a risk of ruin much lower than that, although I could understand if you were willing to risk more than that.
A bankroll of 500 units would provide you with a 10% risk of ruin. I think most recreational card counters who want to get into action would be satisfied with that.
Double Deck Blackjack Bankroll Requirements for Card Counters
Scoblete goes on to suggest having a betting spread of between 1 and 8 units at a double deck blackjack game. This time, he suggests starting with a 2-unit bet and increasing the size of your bets by 1 for every +1 in the running count. He also suggests never betting more than twice the size of your previous bet.
He suggests only betting one unit any time the count is negative. He goes on to suggest that you need a bankroll of 160 units for a single session and 2000 units total.
This advice seems ludicrous compared to what I’ve seen from respected blackjack experts. The overall bankroll size seems to be way too much. The session bankroll is unnecessary. And the bet sizing is also questionable—I don’t know why you wouldn’t base your bet sizing on the true count just because it’s a double deck game.
Four Decks or More Card Counting Bankroll Requirements
For 4 decks or more, Scoblete suggest a betting spread of 1 to 12 units. I think this is an extremely aggressive betting spread. I’ve never even gotten away with a betting spread of 1 to 10 units, myself.
He goes on to suggest starting with one unit and increasing the size of your bet by 1 unit for every number beyond +3 that you get—and this time he’s talking about the true count again. He goes on to suggest that if you’re winning, increase the size of your bets even if the count is negative.
This might work, but increasing the size of your bets when the count is negative reduces your overall expectation. For most card counters, this will have the effect of wiping out your advantage.
He suggests a 200 unit bankroll for a session and a total overall bankroll of 2500 units.
Nope. A 1000 unit bankroll is still plenty.
More Card Counting Advice
This is more general advice that Scoblete offers. I’ve labeled each nugget of wisdom with a “GOOD” or “BAD” flag to let you know which advice you should follow and which you should ignore:
- Get away with a win of any kind – BAD.
- Quit if you get tired – GOOD.
- Don’t chase losses – GOOD.
- Take a nap if you lose your entire session bankroll – BAD.
Get away with a win of any kind. This will result in you quitting a session before you need to. As long as the count is positive, you should keep playing. The ultimate result of advice like this would be to quit any time you have more money in front of you than you started with. That’s neither a fun or profitable way to play blackjack.
Quit if you get tired. This is great advice. A single mistake per hour can wipe out any edge you get from counting cards in blackjack. Tired players make more mistakes.
Don’t chase losses. This is borderline good advice. Most players who are chasing losses are emotional and increasing the sizes of their bets beyond what their bankroll can handle. But if you have your emotions under control, you can rebuy to your heart’s content when you’re losing. In the long run, it doesn’t matter how a single session turns out.
Take a nap if you lose your entire session bankroll. This is just silly. What if you’re not tired? What if you want to play another game? Losing your entire session bankroll is a meaningless measuring stick. If you’re playing in a good blackjack game and need to rebuy, by all means, rebuy as long as you’re confident you won’t make any playing mistakes.
Bankroll Requirements for Basic Strategy Players
Scoblete offers some odd advice for how to manage the size of your bets when playing with basic strategy:
- Start with 2 unit bets.
- If you don’t see any aces during the first round, increase your bet size to 3 units.
- Otherwise, stick with 2 unit bets.
- If you’re winning consistently and are up by 50%, increase the size of your bets to 3 units.
- Any time your winnings drop back below 50%, drop your bet size back to 2 units.
- Try to get away from the table with winnings of 25%.
- If you double your stake, set aside 50%.
- Start with 80 units for your session bankroll.
- Have a total blackjack bankroll of 800 units.
All this advice, by the way, can and should be ignored. If you’re playing with perfect basic strategy, the house has a low edge—maybe 0.5%, maybe 1%.
But they still have an edge.
Raising and lowering the size of your bets in this situation make no sense.
Also, if you’re going to quit after achieving a win goal of 25% of your starting bankroll, you’ll never get to a point where you’ve added 50% to your stake or doubled your stake.
And the size of your bankroll in a negative expectation game only matters in terms of how long you want to play. If you’re playing a negative expectation game, your risk of ruin in the long run is 100%, regardless of how much money you have. That’s just how negative expectation games work.
He also suggests that if you’re losing consistently at one table, you should move to another table. This is more nonsense.
That list of advice above was aimed at single and 2 deck game players. He has more advice for players in 4 deck games or more, but it’s equally useless. To save space, I won’t even get into it here, because it plays on the same fallacies as the other advice.
Craps Bankroll Advice for People Using the Oddsman’s Bet
The oddsman’s bet is a concept in craps I hadn’t heard of before reading Guerrilla Gambling, and I like it.
If you’ve read any of my previous posts about craps, you know that there’s one bet at the craps table with no house edge—the odds bet.
But the odds bet comes with a catch. You must place a pass line bet and a point has to be established before you can place this bet.
Scoblete points out a loophole.
If you can find a pass line bettor who hasn’t taken odds on his bet, you can offer to make an odds bet on his pass line bet. Some people can’t afford the odds bet, while others just don’t want to place the bet because they don’t understand it or whatever.
If they agree to this, you become the “oddsman.” Not everyone’s going to agree to this, of course. Some people will be suspicious. Some won’t understand what’s going on.
That makes sense, too. I’ve never tried this or heard of anyone who has, but it’s worth considering, I think.
From a bankroll management perspective, Scoblete suggests using a “mini-Martingale” here. The example he uses looks like this:
You’re playing in a game where double odds are allowed. The point is 4. You find a player who has a $5 pass line bet on the table. He agrees to let you be his oddsman.
You bet $5 on the odds bet. If the shooter makes the point, you’ve won $10. (The odds bet pays off at 2 to 1 on a point of 4.)
But if he doesn’t make the point, you place a $6 odds bet next time. And so on up to a $10 bet.
If none of those bets win, then you lose $28.
But if any of them win, you win a little bit extra.
I don’t think there’s any drawback to this betting system, although I don’t think it improves your probability of winning, either. Any of these progressive betting systems do the same thing. They slightly improve your probability of having a small winning session, but eventually you’ll wind up with a losing session.
Except when you’re the oddsman. If you continue to play that way long enough, you’ll break even. That’s because the house edge is 0.
He suggests that you have 500 units to play this system, but it doesn’t really matter how much or how little money you use. The system doesn’t increase your probability of winning, but it doesn’t hurt it, either. You could easily be the oddsman with just enough money to place one bet.
Craps Bankroll Requirements for Pass and Come Bets with Full Odds
I might have already pointed this out, but your only logical goal with bankroll requirements when you’re playing a game with a house edge is to maximize the amount of time and fun you have at the table. Scoblete’s advice is okay as far as this goal goes, but don’t be fooled into thinking this advice will improve your probability of winning.
He suggests have 1000 to 2000 total units that you divide into 10 session bankrolls of between 100 and 200 units.
He also suggests sticking with craps games where you can get 5X odds. Your goal is to always get the full odds bet on the table when it’s available.
So if you’re betting $5 on the pass line or on the come, you’d bet $25 on odds. Scoblete also suggests that you always have 3 bets in action if you can, but he admits this is just a preference of his. It has no bearing on the likelihood of winning.
He suggests not increasing the size of your bets if you’re losing. In fact, he suggests lowering the size of your bets if you’re losing.
He also suggests not getting into a hurry when it comes to increasing your bets during a winning streak. He suggests waiting until you’ve tripled the money you have in action before doing that.
And finally, he advises walking away from the table once you’ve lost a third of your stake.
All silly advice, but probably harmless for the most part.
Baccarat Bankroll Requirements
He suggests a 20 unit session bankroll.
If you’re at a $25 table, that means having a session bankroll of $500.
He also advises not staying at the table unless you get on a winning streak.
His advice here is pretty vague.
Pai Gow Poker Bankroll Requirements
Scoblete is a big fan of playing as the banker, but to play as the banker, you need a huge bankroll.
After all, you need to be able to cover the other players’ bets.
He suggests looking at the table maximum, the number of players at the table, and multiplying them.
Then multiply that by 100. Most of us can’t afford that kind of action.
Poker Bankroll Requirements
Scoblete suggest having 100 times the table minimum for each session and enough money to go 12 sessions without winning. He doesn’t provide much guidance beyond this.
Most of the legitimate poker information sites I visit use a number of buy-ins as the guideline for bankroll management, though. These guidelines vary a lot based on what kind of game you’re playing.
For example, if you’re playing in a no limit holdem ring game, you might only need 30 buy-ins. That’s aggressive and assumes that you’re a good player, though. A more conservative approach might be to have 50 or 100 buy-ins. These guidelines work fine for single table sit-n-go tournaments, too.
When you start looking at tournaments with larger fields, you might need as many as 50 buy-ins or up to 600 buy-ins, depending on the size of the field. The larger the field, the more cushion you need, because it’s more likely you’ll bust out of the tournament.
I think it’s safe to ignore Scoblete’s advice about poker bankrolls and look elsewhere for guidance.
Scoblete’s approach to bankroll management is interesting and entertaining, but for the most part, his guidelines are arbitrary in the extreme. Some of his approaches to these games are interesting, but none of them do much in terms of improving your probability of winning.
I’m planning a part 2 to this post, as he covers more games and approaches than I could fit into a single blog post.