Observed house edge in baccarat?

I saw that the Wizard answered a question about theoretical house edge in Macau being roughly 2.85%.

http://wizardofodds.com/ask-the-wizard/276/

Does anyone have any further knowledge around observed house edge in baccarat (proportion of Banker, Player, Tie bets)?

The house edge on Baccarat is a statistical constant with the same drawing rules (about 1% for banker, 1.2% for player side), and the poop on the game is Right here.
The table hold may vary month-by-month (that is, the % of table drop kept as profit), but as a very rough rule of thumb is a 10:1 ratio: so, a long-term table hold figure of 25% corresponds to a game having about a 2.5% house edge (typical of Pai Gow poker), and a 14% table hold (typical of Baccarat with a little side bet action) corresponds to about 1.4% table hold, within limits. Higher holds may indicate a lot of side bet action.

Thanks for the reply.

I'm slightly confused by your use of the word "hold". In the long-term, won't hold converge to house edge? Ex. A roulette table's hold will converge to 5.26% over time even though monthly/daily/weekly variations will obviously be higher/lower?

I'm not quite sure what you mean by 10:1? Do you mean that I should take the highest house edge proposition bet and divide by 10 to obtain the over hold (I certainly wouldn't think this is the case)?

Hold does NOT converge to house edge. It's a function of other factors, such as buy-ins, playing time, hands per hour, etc.

The observed house edge is a weighted average between all the bets people tend to make. There are no playing mistakes in baccarat (except betting the tie...), so they can use a constant estimate based on betting patterns.

Without watching the table carefully and recording all the betting action, there is no way to actually calculate this from the hold.

So you're saying that hold on a roulette table doesn't converge to 5.26% over time?

In games like craps and baccarat, the house edge on the game in entirety is a proportion of the dollars bet on each bet type. My question was if anyone had observed hold that was consistent with the 2.85%.

Quote: JonDough

So you're saying that hold on a roulette table doesn't converge to 5.26% over time?

"hold" measures something else. If you walk in with a Grand and walk out with $200, the Casino 'held' 80% of your money. I believe I have that right. Over time it is averaged for a player.

Quote: odiousgambit

"hold" measures something else. If you walk in with a Grand and walk out with $200, the Casino 'held' 80% of your money. I believe I have that right. Over time it is averaged for a player.

Yes this is correct, and a subtle distinction that took me a while to understand.

The House Edge applies to EACH BET PLACED. Say you walked up to a US roulette table and bought in for $100. If you place a bet for $10, your expected loss on that bet, according to the house edge, would be $0.52. Now if you sit there and place 100 $10 bets, then your expected loss for the session would be 100 * $0.52 or $52.06. That would be the "theoretical hold" for the casino, $52.06 of your original $100 or 52.06% (of course over 100 bets, who knows what the actual hold would be). So "hold" takes into account house edge, size of buy-ins, size and frequency of bets, etc. I would guess casinos use all of those variables to estimate what their hold should be for each table on each given shift.

The confusion may come from this: the hold of a roulette table will converge to 5.26% of ALL MONEY WAGERED, over time. But this is different from the drop (the amount of money the table takes in cash), or the amount of money brought to the table in chips. Actual hold is normally measured against the drop.

Hold = Amount casino keeps / Total amount bet.

The shorter the time period, the more this % will vary. However, on a game like Baccarat or Craps, theoretical casino edge (long-term hold %) is entirely dependent on the proportion of dollars wagered on each bet type. My question was if anyone had evidence to suggest it was around 2.85% for baccarat?

Quote: JonDough

Hold = Amount casino keeps / Total amount bet.

No, this is incorrect. Hold = Amount casino keeps / Total value of chips purchased. If someone buys $1000 in chips, then goes directly to the cashier to cash them out, the hold for this person would be 0%; this will skew the observed hold, and the person didn't even wager anything.

Quote:

The shorter the time period, the more this % will vary. However, on a game like Baccarat or Craps, theoretical casino edge (long-term hold %) is entirely dependent on the proportion of dollars wagered on each bet type. My question was if anyone had evidence to suggest it was around 2.85% for baccarat?

I think we've been distracted from your original question.

Yes, the actual house edge depends on the proportions of each bet the players are taking. It's only mathematically possible to determine the exact house edge on a single bet at a time. The Wiz proposed a proportion of bets that would match with what the Macau casinos are reporting for the investors. But the actual proportions for each bet and thus the combined house edge is something that would vary from casino to casino, city to city, and would require a lot of observation and data collection. I'm certainly not an expert on the gambling industry, so I don't know if that kind of data exists or would be available. Based on the people I've seen playing Baccarat (very superstitions, try to play streaks, will bet the Tie if it hasn't come up in a while), the 2.85% they use in Macau doesn't sound too far off.