1 out of 1 members found this post helpful.
Did you find this post helpful?
Yes |
No
Getting a handle on bad variance
First of all I'd like to say thanks to all those on this forum who have taught me so much. I've been counting on a recreational basis for almost a year now but I continue to learn new things almost daily. I post only occasionally but I visit the site several times a day. I'd like to especially thank Norm for hosting this great site and for his phenomenal CV software (which is a must have for anyone who wants to count). Enough ass kissing ![Beguiled](images/smilies/beguiled.png)
People often warn newbies about the enormous swings inherent in this game but I think it's difficult to really get a grasp on the likelihood of a significant downswing. Many focus on the hourly win rate and when they hit a bad stretch, find it difficult to accept/believe. I've recently hit a bad stretch myself and it got me questioning whether I'm actually playing a winning game or not. Below is some information I've tried to remind myself of during my recent soul-searching. I think this is all correct. Feel free to add or edit as you see fit:
Wherever your current bankroll sits, it is more likely than not to go down, at least somewhat, in the short term but is still likely to go up in the long term. Regardless of the count, you're more likely to lose the next hand you play than you are to win it. Consider Table 84 on page 197 of Wong's Professional Blackjack. A HiLo player spreading from $5 to $50 on a 6D shoe who sits out negative counts and plays 4,000 hands has a 50% chance of going at least $900 in the hole at some point during those 4,000 hands. If he's a green chipper spreading from $25 to $250, he has a 50% chance of going $4500 in the hole at some point.
I use CVCX to get my own "50% chance of losing" number and I'm curious to know if what I'm doing is correct. After I put in my data, I then click the "Freeze custom bets" button and start decreasing the bankroll number until the risk of ruin hits 50%. I've been surprised how high this number is! For instance, if I plug in HiLo Sweet 16 Fab 4 for 6D 4 players H17 DAS Sr 75% pen 100 rounds/hr 1:12 spread $25 min play all 30K BR, I get an hourly win rate of $50.74 - not too shabby. If I then freeze the custom bets, and decrease the bankroll I find that I have a 52.9% chance of being down at least $7,725 at some point. That's more than 25% of the bankroll. Now admittedly that assumes no Wonging, but still, that's scary when you know that's PERFECT PLAY with no cover. For my particular game, my 50% chance of losing number is about $6,000.
This not only applies to when you start, it applies to any time in your career. Take me for instance. When I first started, I went down $1600 before I started winning. I then hit an all-time high of $4,500 after 50 hours of play (averaging $90/hour - well above my expected EV) and thought I can't lose. Well, I can and I did. Over the next 35 hours, I lost $4,100 (including 6 straight losing sessions recently) and I'm now almost back to even. In fact, when you consider transportation costs (which I track), I'm actually losing money. The scariest thing is this - despite having lost big recently, I still have a 50% chance of losing 6K more. That 4.1K I've lost recently doesn't mean my "50% chance of losing" number is now 1.9K. Nope, it's still 6K. There's no such thing as being "due" because you've lost recently. The cards don't remember that you just lost your shirt.
Card counting may be simple, but it sure ain't easy!
Bookmarks