One reasonable way to look at these decisions is to figure out how many responses could plausibly be misplayed by a weaker player against either play. For example, with an opening roll you gain bonus equity by splitting with 5-4 because many opponents will incorrectly fail to hit with a 5-3 response. Here, just to stick with that specific example, everyone is hitting with 5-3 after you split so it doesn't matter. But if we spent some time with this maybe we could figure out how many errors you are likely to induce with each play.
Inducing errors in response is a worthwhile and valid concept (and can be used against strong players too), though it's a different issue from increasing/maintaining contact for greater long-term complications.
Nevertheless, I'd like to address your error-inducing example:
Faced with 54S-53 or 54S-62, I can't remember anyone making any play other than H (Hit, 13/5*) against me in the last twenty years. Apparently your experience is different and I'll take your word for it that people do. Even so, IMO, it's a lot more likely that someone will misplay 61P-54S-53 or 61P-54S-62, simply because the positions arise with relative infrequency -- he is less likely to copy it from remembering having seen experts do it. Moreover, there is value in replenishing the spare on the stripped 8pt/7pt (that 61P-54S-53/62 has and 54S-53/62 does not).
Consider the position below. Do you think it is automatic that players will hit with 61P-32S-63 or 61P-32S-54? I don't. In fact, as I recall hitting is an error with both rolls! And yet "everyone" does (and should) hit with 32S-63 or 32S-54 (same positions without Blue having made the 7pt).
| |
|---|---|
| Blue rolls 63 or 54 | |
In no way do I mean to discourage you from further response-error estimation on this position or any other; on the contrary. But it's a tricky subject and even experts will not always agree on the likelihood of human opponents misplaying various rolls.
Nack