| |
BGonline.org Forums
Non-doublet Replies: Rules for beginners
Posted By: nack ballard In Response To: Non-doublet Replies: Rules for beginners (Axel Reichert)
Date: Saturday, 29 April 2017, at 8:18 a.m.
I need to get one thing out of the way. Jeremy defined the full-roll definitions of D and S correctly (at least in the beginning of the article, the only part I've read). Where you went wrong was in making up your own rules for partial rolls (that necessarily have restrictions). In Nactation, you should regard partial-roll usage as extra-curricular, just a bonus when you can use it to your advantage (e.g., see mention of 61D later).
You seem to be carefully eliminating plays that a certain data set tells you are between .06 and .07 (or .08?) without knowing if they really are worse than .06, and you seem satisfied to include plays that are in the .05 to .06 range without being sure they aren't worse than .06, and locking in your model before you know. I hope you understand the potential arbitrariness of that. However, I won't bring it up again (unless you do).
You might want to state somewhere, if you haven't already, that these replies stand up against the moves in your opening library, not necessarily against unorthodox openers. And while we're on the topic, I've been assuming the library is the set of 27 openers I listed. Or perhaps 29 with 62R and 62$, which shouldn't change anything, but in any case not against 43U.
I like your current usage of "a" and "b" much better than the way you used it previously (where what preceded "a" was arguably another "a").
You are right that spelling out the phrases is easier to understand, but I don't think it's easier to recall. I recommend that you keep the longhand version (this one or what you later change it to) -- call it version "1", but then also make available a "version 2" that seriously cuts down on the wording so that once someone understands version "1" they can upgrade to the easier-to-memorize "version 2." (Nactation might be used in Version 2, or perhaps incorporated into a version 3 that is even shorter.)
I understand your objection to "else." I think careful punctuation eliminates or diminishes the chance of misinterpretation (and at least you have to acknowledge a tradeoff), but if you don't like it for version 1, it should be fine for version 2.
If you're going to take the trouble to write everything else out, consider (instead of "hit > makeable point") writing "hit higher than makeable point" or "hit above makeable point." On the other hand, when preceding 0.06 or 0.08 in the parentheses, I think the > is fine, looks more like an arithmetic expression.
Likewise with "Hit > 3. Hit 3 with 3." Clever, yes, but I urge you to save it for version 2. How about, "Hit above the 3pt. Hit on the 3pt with a 3." BTW, in general, I think the investment of "pt" (the expenditure of merely two small-case letters to define a point number) is very wise, even though it is not possible to hit 3 checkers with non-doublets. That said, I think "do not hit 8/3*" is not only stronger but also more efficient.
In 2a, why refer to "two outer boards"? If one hits in the far outer board, there is no roll portion left over. And in 2b the "two inner boards" cover everything else, so why not just say "elsewhere"?
I'm not sure where you're drawing the line with Nactation. Evidently, you're not confident 43Z will be understood, but if so then why should the reader know what "split reverse" means? Indeed, why should he know that "split" means to split with the larger die and come down with the smaller die, and if so then why not use the shorter "S"? If you decide you want to be entirely Nactation-free (though I don't recommend it), it seems like you'll have to replace both split clauses with something like "...split with larger die (exc. prefer 21pt) and come down with the other die."
Personally, I think it's better to point the beginner to Section 1 of the Nactation tutorial (if you're not there to explain in person), and have them read what the five basic symbols mean (mostly diagrams, only about 1/3 page of real text), plus Z (top of next page). Once they understand the meaning of "S" and "Z," then they will also know that 42P-32S and 42P-32Z mean the same thing (i.e., no, my rhyme didn't mean 32D). That, along with all the other benefits.
I think "If move is blocked, play most similar legal alternative" assumes too much. For 64P-41 (for example), that would cause me to move 41B or 41W, in order to play in(to) the same quadrants as with 41S or 41$. If you want the reader to confine himself to the opening library I think you should state "if blocked, use a move from the opening library." (And that addresses your astute point about 64P-x1$, though not about 61P-65D for which I'll add D at the end unless you have a better idea.)
On to the algorithm itself.Should exceptions be listed before or after the fact? If they are handled individually with "except" or "unless" in close proximity, then after is fine. But if the exceptions are handled as a group (which I do think is best), then I think BEFORE is better. The reason is simple. There is an implication of hierarchy (albeit with a hit / no hit branching). The brain is okay with switching the order of two things next to each other, but otherwise one can feel betrayed.
Let me state this point more clearly. As it stands now, you don't have the luxury of stopping when you get to the roll you want to play. For example, rule 2b tells you to play 41$-64Z. But when you read to the end (IF YOU DO) you discover that you're actually supposed to make a different play. I'd rather see the exceptions listed at the top. Whether or not you officially number it as a rule you can entitle it "Learn these first" or whatever.
You did a terrific job on the pointing / non-pointing and hitting / non-hitting structure. Too bad about all the extra words, but it just seems to hard to clearly convey the hierarchy otherwise.
Unfortunately, I believe you need to also state WHERE the reader should hit (e.g., consider 32S-62), and I think that means on the highest point possible. (This is one advantage "H" has over hit. Hit on highest point is already subsumed in the definition of the letter, though I understand you might not want to use H for other reasons.)
Here's what the set looks like with my proposed changes:
Replies for playing without blunders (> 0.06)Non-doublets
1. Cardinal rules:
....a) D-x1$ .......(Slot with ace after Down)
....b) D/$-64R ..(Run with 64 .vs. Down or Slot)
....c) 6xS-65K ..(Kill with 65 after 6-Split)2. Hitting
....a) With pointing roll: Hit above makeable point.
....b) With other rolls: Hit on 3pt (exc. with 8/3*) or higher.
........(Always hit on highest possible point.)3. Hitting supplement (if leftover die)
....a) After hitting in the outer board: Play up.
....b) After hitting elsewhere: Play from mid
........(If "up or "from mid" is blocked, do the other.)4. Not hitting:
....a) P, 65R, 43Z, S.
........(If blocked, play one of the 27 openers or 65D.)
I've run out of time. I'll look at the doublets later.Nack
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.