home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: rec.gambling
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!cs.utexas.edu!hellgate.utah.edu!hellgate!jacobs
- From: jacobs@cs.utah.edu (Steven R Jacobs)
- Subject: Re: Splitting 10's (was Re: LV recommedations requested...)
- Message-ID: <JACOBS.92Nov20073601@cells.cs.utah.edu>
- In-reply-to: hall@kronos.arc.nasa.gov's message of Fri, 20 Nov 1992 03:24:17 GMT
- Organization: University of Utah CS Dept
- References: <1992Nov9.201259.644@fibers> <1992Nov10.021330.12874@kronos.arc.nasa.gov>
- <JACOBS.92Nov9205337@cells.cs.utah.edu>
- <1992Nov20.032417.5139@kronos.arc.nasa.gov>
- Date: 20 Nov 92 07:36:01
- Lines: 41
-
- In article <1992Nov20.032417.5139@kronos.arc.nasa.gov> hall@kronos.arc.nasa.gov (Michael Hall) writes:
- >In article <JACOBS.92Nov9205337@cells.cs.utah.edu> jacobs@cs.utah.edu (Steven R Jacobs) writes:
- >>In article <1992Nov10.021330.12874@kronos.arc.nasa.gov> hall@kronos.arc.nasa.gov (Michael Hall) writes:
- >
- >>> Preliminary simulation results indicate that it costs close to 20%
- >>> of overall profit if you fail to split 10's.
- >>
- >>20%? I'm extremely skeptical of this claim. You are saying that a card
- >>counter with a 1.5% edge will have a 1.8% edge by adding just this one
- >>strategy adjustment? I really doubt it. I might believe that correct
- >>insurance plays would be worth that much, but not splitting 10s.
- >
- > Sorry for waiting a while to respond...
- >
- > Someone emailed me saying he found not splitting 10's will cost 17%
- > of expected winnings; as I recall I found that it costs 17% of return
- > on investment, but I can't find that old article. So a 1.8%
- > bankroll-per-hour-return-on-investment counter would have a 1.5%
- > bankroll-per-hour-return-on-investment just by dropping this one
- > adjustment.
-
- I'm sorry, but these numbers simply do not sound reasonable. A player
- will get a pair of tens about one hand in ten. These will be up against
- a 5 or 6 only one time in 6.5 hands. So, one hand in 65 will be a
- "candidate" for splitting tens. The count has to get fairly high in
- order for ten splitting to be profitable, but I'll be generous and say
- that 20% of these hands will be splittable (this figure is probably way
- too high). That means that one hand in 325 will be a pair of tens that
- are worth splitting. Now, if these splittable hands give on average a
- 10% edge to the player (compared to standing), and these are always played
- with a 5 unit bet, then you will gain an extra half unit every 325 hands.
- With a 1.5% advantage, the card counter can expect to win about 4.875
- average bets in these same 325 hands. So, if the average bet size is
- about 2.something units, the player wins about 10 units without splitting
- and about 10.5 units with splitting tens. This is about a 5% increase,
- and I think I've been overly generous with my numbers.
-
- In short, 10 splitting just isn't that valuable.
- --
- Steve Jacobs ({bellcore,hplabs,uunet}!utah-cs!jacobs, jacobs@cs.utah.edu)
- "Don't worry, I just have these harmless pocket rockets...."
-