The case for Collins

[bricap]

I have just spent nearly two weeks playing 72 games of Collins. I'll spare the details of my performances in the two respective tournaments. Nothing to really write about there, except that I probably will need three years to master the Collins and the intricacies of the game associated with that book. Cramming for three months before one of

Comments

[nigelbo]

I am convinced. I will try to get around to it sometime soon hopefully, but I would like to support the cause by actually learning and giving it a try at some point for sure.

[wallydraigle]

First of all, I have enormous respect for Brian's strategic prowess within the game of Scrabble, and his contributions to Scrabble theory. I think this is the most articulate case for Collins I’ve seen in a while. Watching Brian's annotated games has undoubtedly made me a better player. And I understand Brian's desire to take up Sowpods after Causeway. After all, Brian has a much stronger understanding of TWL than of Collins. I’m not trying to sway Brian into not playing Sowpods, but I’m just trying to state the case for TWL

[jigsawn]

There isn’t a lot of skill in 550-500 contests eitherThere's the skill of learning, retaining, and finding those words. There are many skills that go into Scrabble besides those of strategic thinking. I don't mean to denigrate strategy at all, but it's not the end-all be-all in this game, and you seem to be implying that it should be

[jigsawn]

strength OR a weakness, that should have said

[evwhore]

Sowpods rewards mindless equity maximization a lot more than TWL does.

I'm not sure how you're defining equity, but whatever it is, it's wrong, so please stop using the term this way :-) You mean something else, so choose a different word...

"Equity" has a specific game theory/probability meaning, and since you're an influential Scrabble thinker, I'd rather not have lots of people learning a distorted definition of the word :-)

[jigsawn]

Equity in this context means:

(score of current play) - (expected score of opponent's next play) + (expected score of next play)

Essentially, it is what is calculated in a 2-ply sim. There are good reasons for going exactly 2 ply ahead. In this way, you're taking into account the defensive value of the play and the value of the leave toward your next play. Anything beyond that is much less meaningful, because the tiles on the racks and the shape of the board are so unpredictable.

Equity is a very useful concept for Scrabble analysis and it is accepted jargon in the community, which is a nice way of saying buzz off and stop telling people they're wrong when you don't know what you're talking about.

[evwhore]

buzz off and stop telling people they're wrong when you don't know what you're talking about.Gosh, thanks for saying it the "nice way" I guess

[sowpods]

Dave, a more proper term for the quantity you have listed would be "effective score" or "adjusted score". (See this page for an analogous idea in CS: http://en.wikipedia.org/wiki/Potential_method) Evwhore is right in noting that this isn't "equity".

[nigelbo]

It's hard to construct a proper reply when teh initial writer clearly has no clue what theyre talking about. Sowpods is obv a super deep game with a lot of thinking behind it, and the failure of most good-not-great players comes from exactly what is quoted, simply playing high equity plays, or trying to emulate quackle in effect, with some minor modifications

[ar_raqis]

"It's hard to construct a proper reply when teh initial writer clearly has no clue what theyre talking about."

Well, yeah, obviously. You (David) have played 422 WESPA games (winning consistently) and Kenji has played 0. And I would think Kenji would realize the importance of that. He understands TWL quite deeply, but I'm very doubtful that he understands the Collins metagame well enough to be trusted on the sort of high-level strategic points he's making.

Also, Kenji is sort of on an ice floe in the middle of nowhere with respect to his idea of what Scrabble should be like. Your crack about removing the blanks made me laugh, because that is exactly the sort of thing he would suggest -- he once proposed that the penalty for an unsuccessful challenge should be forfeit of game.

[bricap]

Let's not turn this into a credentials debate. Kenji is making arguments that will be made by others after I take this to cgp. Right now I'm not sure if Kenji was trying to dissuade or if he was trying to play devil's advocate. He is a philosophy major, after all

[wallydraigle]

My OP wasn't meant as a complete rebuke of Sowpods, but as a defense of TWL in terms of Sowpods being the "ultimate game". TWL still matters. It's pretty insulting to me and to the game in general when so many top players undercut the dictionary that we use

TWL is too easy

[nigelbo]

I think that Nigel is a good example of how easy the TWL game is. The last three times at the Nationals coming in 1st, 2nd and 1st. And a crushing four wins clear this year. No such domination is evident in Collins games. Collins is a harder game to master and hence is more challenging. Consider whether you really do want to stick with the easier TWL? TWL is less dynamic, less dramatic, less interesting and less strategic. Yes, I said less strategic. Collins games can still be blocked out (just as with TWL) but it requires much more skill to do it properly. And it requires a lot of future thinking and planning to decide whether or not to try to block out the opponent, or whether to go for the run away win. If you get it wrong then you lose. In TWL the option is almost always block out the opponent because it is demonstrably the much easier option and requires less effort. Everyone likes to choose the easy blockout option when ahead, and everyone gets frustrated by their opponent choosing the blockout option when behind. Very often

Re: TWL is too easy

[nigelbo]

ah, much better composed than what I wrote mr anon., unfortunately a lot of people will refuse to accept this until they play and master the game themselves. The crazy thing is the strongest argument isn't even that it's a superior game strategically (prob), aesthetically and flow-ing-ly, its that America willingly (?) segregate themselves from the rest of the world, which not only works negatively for the rest of the world but works negatively for in particular AMERICA! It seems only logical that once the Americans join the rest of the world only positive changes will be forthcoming... More prizemoney, more people on the international scene (and vice versa), more popularity for the game, and increased growth. Just to name some. Having a SINGLE community to make decisions about the Rules, about the tournament draw styles, about increasing publicity and game promotion, and about what color hats and vests are best to wear while playing.
you get the idea. the possibilities are endless. just some food for thought.

Re: TWL is too easy

[bricap]

The game has changed even since Ganesh left. This year's Causeway was the toughest field for that event, yet. Some of the players he faced in those Causeways didn't make the Premier group this year. He'd do well today, I'm sure, but that kind of track record would be pretty hard to come by at this point. You'd see similar dominant runs by top players in our game way back when that just don't happen quite as often today.