SP21/Pontoon Bet Spread Sizing?

[Bushie]

My research into SP21 says to the contrary. Covariance is high in high playing counts compared to low playing counts and low when my balanced ace side count is high (this is different than surplus and deficit aces). As for the combined betting count covariance is higher with when the count is higher.

Now for bet sizing, which is extremely accurate with my 2 count system, it is almost always better to bet 2 spots. Of the 260 TC pair bins that aren't simply catchall max bet or min bet bins 2 of them are better bet with 1 spot, the other 258 are better bet as a 2 spot bet. There isn't much difference until you get to large bets which are much better bet as 2 spots. Below are the averages that would be bet at each betting level and the fraction of the optimal 1 spot bet bet optimally on each of 2 spots.

AP unit bet made at 0.5% advantage: not betting optimally at this point so irrelevant.
Second advantage bet size: 0.76%
Third advantage bet size: 0.79%
Fourth advantage bet size: 0.80%
Fifth advantage bet size: 0.77%
Sixth advantage bet size: 0.81%
Seventh advantage bet size: 0.83%

Disadvantage bets would be bet optimally with a bet of 0 so this is also irrelevant.

For 1 spot to be the better choice the ratio of the optimal bet on each of 2 spots to the optimal 1 spot bet would need to be 0.75% or less. The sims say that per card eaten the results are fairly equivalent for playing 1 spot or 2 when you take into account that you get 50% more rounds out of the shoe heads-up with the dealer playing one spot rather than 2 spots. Game speed would be the determining factor. I have had dealers that have a long pause between rounds where 2 spots would clearly play through a shoe faster than 1 spot. For dealers with no lag I am not certain whether playing 1 spot or 2 spots would play through a shoe faster. I think 2 spots would play through the shoe faster if the dealer played his hand out every round but since he won't have to play his hand out more often if you play 1 spot that might make 1 spot faster. It might also get an extra round or 2 in rather than simply 50% more rounds which would also help make up the small difference in optimal bets per card eaten that favors playing 2 spots to get more money out per card eaten and therefore more money out across the entire advantage situation.

3 spots bet optimally is very rarely best heads-up. Again it is not best by much. Most gains in optimal betting are less than the error range so there is no clear winner despite the stats saying you are almost always best off playing 2 spots for less than half of the time. The bulk of the time 2 spots is a clear statistical winner.

Of course all of this is count dependent so your count may have different results.

So the difference between 1 spot and 2 is so marginal that it hardly makes a difference? And disadvantageous bets are optimally 0 aka wong out?

If thats correct you're both giving me contradicting analysis here. I'm not sure what to think..

[ZenMaster_Flash]

"I'm not sure what to think."

I would pay very close attention to what Tthree posts.

[Bushie]

I would pay very close attention to what Tthree posts.

I have. Granted, a lot of it is over my head, but surely higher covariance in higher counts is a bad thing? Wouldnt that outweigh playing 1 spot with a 50% smaller bet?