Saturday, April 10, 2010

What negative expectation really means is that the house has a very poor opinion of you and expects you to play like an idiot. Surprise!

_
(No 7-dog Trial picks for Sunday, April 11)

The conventional wisdom is absolutely unequivocal: If you risk money at any casino game of chance, you will ultimately lose.

Some experts try to flex their math muscles by predicting that if you belly up to a game with a 3.0% negative expectation, you can expect to lose 3.0% percent of your total action.

So if you bet an average of $100 against 100 consecutive betting options, your action (or handle or churn etc.) will be -.03 x $10,000 = -$300.

Impressive, huh?

And totally wrong.

Unless you bet a fixed amount every time, take no defensive action at any point, and the house edge at the end of 100 bets precisely matches the game's overall negative expectation.

None of those things actually happens in real life.

The house's expectation when you buy chips at a table game is that you will over-react to both winning and losing streaks, choosing your bet values randomly and risking more when you are losing than when you are ahead.

If you get ahead of the game, you will press your luck (again, randomly) until eventually the house edge sets in and whittles away at your puny profits until your chips are all gone (or are, at best, seriously depleted).

You will fail to exploit winning patterns, and will panic when a prolonged losing trend comes along.

You will, as I said before, play like a twerp.

As I reported earlier, I am gearing up to do much more betting online in response to worrying developments at my local casinos.

Single deck 21 with loose rules used to be everywhere, but now, most of the big casinos severely limit splits and double-downs/re-splits/redoubles at those layouts, and have slashed the payout on a natural from 3:2 to 6:5.

The minimum bet in the big casinos is now $10 except before lunch and past my bedtime.

And layoffs have so greatly reduced the number of open tables that seats are often hard to find, and a crowded game means slow progress against drunks, fools and other hazards.

Heads-up (solo against the dealer) is the best way to play blackjack, but these days it is an impossible dream outside of the graveyard shift.

So, Pinnacle and competitors, here I come.

I have been testing the waters at Pinnacle (with whom I promise you I have no connection beyond the few hundred bucks I just wired into my account!), and my mathematically impossible winning streak continues unabated.

It is possible, I readily concede, that the online casino's so-called practice game is a whole lot easier to beat than the one that kicks in when real money is on the line.

But the fact is that the house edge currently stands at -1.9% and I am notionally more than $20,000 ahead.

That house advantage is a little on the high side for perfect basic strategy play against a multi-deck shoe, but the Pinnacle software shuffles the "deck" after every round so the number is not so surprising after all!

Since I don't count cards (tried it for a while, hated it) the constant shuffling trick does not bother me.

I don't know exactly what my action to date has been, but after 1,887 bets, my educated guess would put it at about $75,000.

According to my records, I have been placing an average of about 250 bets per hour of play, earning around $2,850/hr.

Pity it wasn't real money!

But to repeat the unequivocal mantra of those "experts": right now if their math was accurate, I would be -.019 x $75,000 = -$1,425 in the hole.

Oops...

If you blackjack players out there have not tried online betting before, give it a shot, using the following modification of standard betting rules, tailored for a low-bet game ($1 minimum, $500 max):-

Bet $1.
After a win, bet $2, then $5, then $10, then $15, $25, $40, $60, $90 etc.
After a loss, bet $5, then $10, $25, $25 etc.
After a push, increase the bet to the next "win level" ($5, $10, $15, $25) but do not bump the bet once you reach $25.
Make $25 your increment once you get that high, and use it to calculate your target after a mid-recovery win (Example, -$1, -$2, -$5, -$10, -$25, -$25, +$25 requires a next bet or NB of $75 to recover 2 units +1u).


Play strict basic strategy, grabbing ever split and double-down the game will permit, and you should end up with results that look something like this:


The grey area in the summary above highlights a departure from my standard stop-loss of $2,500, or 2,500 minimum units.

The one and only time I got into serious trouble, I opted to drop out of the low-rent game and fire up Pinnacle's high stakes option, which allows bets up to %5,000.

As expected (by me, not by the house), recovery was very quickly achieved once the table limit was hit.

At no other point in the 54 short blackjack sessions to date has the stop-loss been reached, and by the start of session #24 above, I was $10,580 ahead and well able to accept crying Uncle at -$2,500.

Count the grey lines as a stop-loss setback, and the total win to date drops from $21,000 to $17,460.

If Pinnacle delivers that kind of profit when real money applies, I will be a very happy punter.

Friday's 7-dog trial picks gave us just two piddly wins out of seven, costing the 50x tracker $3,735 and 5x $1,275 it could ill afford.

Again, the lesson we take from this trial: spread wide, spread wide, spread wide!

Today I am backing the NBA's Boston Celtics (#509) and Golden State Warriors (#515), both at +110, and baseball's St. Louis Cardinals (#955 at +125), Houston Astros (#958 at +115), LA Dodgers (#959 at +115), San Diego Padres (#961 at +135) and Tampa Bay Rays (#972 at +120).

Truth is, it doesn't much matter what I pick: In the end, the numbers will deliver a profit, in spite of a few bumps along the way.

It's worth noting that the 50x trial once fell $15,000 into the hole but now stands at +$48,660 and has not been in red-number territory since December 15!

The 5x rules set has been bouncing up and down in the red since January 27 after hitting +$6,000 on January 18, 80 days into the 7-dog trial.

Will 5x ever recover?

Probably not. Just as a blackjack or baccarat player spreading a tight 1 to 5 can expect to go broke in short order.

An important reminder: The only person likely to make money out of this blog is you, Dear Reader. There's nothing to buy, ever, and your soul is safe (from me, at least). Test my ideas and use them or don't. It's up to you.
_