Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

An important but unnamed class of positions?

47 views
Skip to first unread message

Tim Chow

unread,
Jan 31, 2014, 1:50:20 PM1/31/14
to
In the process of trying to devise a coherent filing system for my growing list of interesting positions (mostly positions where I erred), I'm starting to think that there is a class of positions that comes up frequently but that doesn't have a name.

Some positions, of course, are easily identified as blitzing or priming or backgame positions. However, often the middlegame does not crystallize into one of these structures. I am starting to find it helpful to categorize this class of somewhat "amorphous" positions according to how many checkers back each player has:

- 0 checkers back
- 1 checker back
- 2 checkers back on separate points
- 2 checkers back, anchored.

By "back" here, I usually mean on the 18pt or further back (which also happens to be the dividing line between U and R, for those familiar with nactation). In principle this gives rise to 10 different subclasses of positions, but only four of them have names (I'll use "a" for "anchor" and "2" for "2 checkers on separate points"):

0 vs 0: race
0 vs a: holding game
a vs a: mutual holding game
0 vs 1: containment

In a recent post I mentioned how I was discovering that "1 vs a" positions had some unique characteristics, e.g., the "a" player is focused on containment and often does not take the risks that Magriel's criteria might suggest. I think that similarly, each of the remaining 5 classes has its own important characteristics. It's also useful to consider all 10 classes as a group because they can transition to each other and the strategy often centers around making a favorable transition.

For example, in a "0 vs 2" position, it's often a key decision for the "2" player whether to anchor up. The "0" player has to decide whether to attack and risk dropping down to a "1 vs 2" or "1 vs a" game.

Or in a "1 vs 1" or "1 vs 2" position, the "1" player often has to decide whether to make a bold Pottling play in order to convert into a (usually) more favorable containment or holding game.

Or in a "2 vs a" game, the "2" player may have to choose between converting to an "a vs a" or a "1 vs a" game.

Even some classical decisions can be framed this way. Anchor-breaking, for example, often concerns transitions from "0 vs a" to "0 vs 1" or from "a vs a" to "1 vs a".

I'm not yet sure whether this is the best way to think about these positions, but it does seem to be a useful way to organize (for example) what is currently a very disparate set of "Pottling" positions that I've collected. I may have some concrete examples to share later when I feel I understand things a bit better.

---
Tim Chow

check...@yahoo.com

unread,
Jan 31, 2014, 5:28:57 PM1/31/14
to

I know there are some classes of positions that are tough to put a name to but a lot of the ones you seem to list here you've already put a name to. I don't see any use in trying to break them down further like it appears you may be doing. For example I give a couple of lessons on checker play in holding games. The checker decisions (for the person in the holding game) are when to run/break anchor or when to take a higher anchor. (typically the defensive bar point) I don't see the use or point of labeling that a 2 or 3 back v. 0 back position. They already have a decent label and they're already broken down so I can understand the themes that arise. Referring to a 20pt anchor game when you might want to run off the anchor as a 2 back v. 0 back switching to a 1 back v. 0 back seems muddling/confusing to me. Idk, let me know if I have it wrong.

Stick

Tim Chow

unread,
Jan 31, 2014, 7:52:17 PM1/31/14
to
On Friday, January 31, 2014 5:28:57 PM UTC-5, check...@yahoo.com wrote:
> I don't see the use or point of labeling that a 2 or 3 back v. 0 back
> position. They already have a decent label and they're already broken
> down so I can understand the themes that arise.

For the ones that already have a name, the label is not that useful, I admit, except to show their relationship to the ones that don't have a label. The main point is to try to categorize the ones that don't have a label.

I hope to post some actual positions in the not-too-distant future, which I hope will clarify my point much better than any words could do. But in the meantime, let me say this. I've amassed a bunch of positions where the player whose turn it is has one or two checkers back, not anchored, and has to decide whether to run into the outfield, or play something else---perhaps bring down a checker or two from the midpoint. Initially, I labeled the one-checker-back positions "Pottle." This already is helpful, I think. What did people call these positions before Pottle stated his "law"? Nothing in particular, I assume. It was a helpful advance just for Pottle to articulate this rule of thumb and notice that it was a good move more often than people thought.

What I've found, though, is that the "Pottle" positions are very diverse. To make sense of them, I feel that I need to subcategorize them further. In addition to the usual things like what the race is and what kind of attacking and blitzing potential each player has, I am, at least preliminarily, finding it useful to divide into cases depending on how many checkers the opponent has back.

To put it another way, don't you agree that it's useful to have the words "holding game" and "mutual holding game" to describe certain kinds of positions? They help organize one's understanding of the game. We don't, however, have a name for a position in which each player has a single straggler, or where one player has a straggler and the opponent has a high anchor. I'm suggesting that giving these situations a name may be helpful.

---
Tim Chow

check...@yahoo.com

unread,
Feb 1, 2014, 6:35:55 AM2/1/14
to

I'll have to see your positions before I can comment further. Yes, it's useful to have holding game or mutual holding game but the other types of games you mention, single straggler v. single straggler or single straggler v. high anchor, aren't all that interesting in general. Run the guy, end of story. Yes, window for you to show the exceptions...but really, run the guy.

Stick

Tim Chow

unread,
Feb 1, 2014, 12:47:46 PM2/1/14
to
On Saturday, February 1, 2014 6:35:55 AM UTC-5, check...@yahoo.com wrote:
> but the other types of games you mention, single straggler v. single straggler
> or single straggler v. high anchor, aren't all that interesting in general.
> Run the guy, end of story. Yes, window for you to show the exceptions...but
> really, run the guy.

I agree. But this gets you only as far as "make the five point" or "always run the back checker." Good as a start, but not enough to make further progress. My "Pottle" subfolder is comparable to my "five point" subfolder at this point.

Also, "run the guy" doesn't apply to the player with the *anchor*. The containment strategy for the player with the anchor is more subtle than I expected before I started studying it specifically.

Finally, you've only mentioned checker play. There is not much out there in the way of cube reference positions for these kinds of games.
---
Tim Chow
0 new messages