3 ms·
Been casually trying to increase my ELO on Wordfeud. When you know the dictionary by heart (Wordfeud has rather restricted dictionaries) and have the basics of
by gillesjacobs 4y ago
Been casually trying to increase my ELO on Wordfeud. When you know the dictionary by heart (Wordfeud has rather restricted dictionaries) and have the basics of bonus hunting and hooking down, Scrabble becomes a pure game of imperfect knowledge and strategy. The known information is the prior letter distribution, the played letters on the board and in your rack. The unknown information is your opponent's rack. However, you can compute exact letter probabilities for his rack and strategize accordingly.
Then the game comes down to managing your rack and your opponent's.
Here's are some common techniques/strats:
- Q-pinning: blocking bonus squares or hooks by using difficult letters so your opponent can't lay down a high scoring turn.
- Q-burning: the inverse, taking a hit on current turn score just to get rid of difficult letters.
- E-burning: sometimes having all-low score letters in your rack is worse than having Q-letters as this will hamper your ability to get high scores on good board positions. A well-balanced rank is key to winning.
- Q-swapping: late-game strategy, maintain a difficult letter in your rack to swap in one of the last turns so your opponent draws the impossible letters forcibly. Remember unused rack letters will be deducted from final score. This dirty trick has won me many close games.
- bena 4y agoTangent: Elo, not ELO. The rating system is named after the creator Arpad Elo, it doesn't stand for anything.
- hkgjjgjfjfjfjf 4y agoTechnical correctness, the best sort of correctness
- bradknowles 4y agoWith respect, ELO stands for Electric Light Orchestra.
- Someone 4y ago> However, you can compute exact letter probabilities for his rack and strategize accordingly. I think the best programs run Monte Carlo simulations on the contents of their opponent’s rack. That includes modeling their opponent’s moves, and thus improves on just computing these probabilities. As a simple example, if they played a move, but, if they had an ‘e’, could have made a much better move, chances are they didn’t have an ‘e’ when they made their last move, even if that, a priori, was still highly likely.
- zem 4y agoI don't think either quackle or macondo do opponent leave inference (elise might though), they just run forward simulations using the pool of unseen tiles.
- jasone 4y agoThis was the idea that I kicked around for years before eventually writing a Scrabble engine in 2018. But by the time I got to the point where Monte Carlo simulation of the opponent's rack leave became feasible, I questioned the utility. Part of the problem is that if your opponent knows you incorporate such a simulation into the analysis, it's possible to subvert the analysis. And the most obvious counter is to not do the simulation. Another issue is that the advantage gained by opponent rack leave simulation is subtle, and it's costly to run enough games to measure the overall utility. I ended up instead putting my effort into controlling search depth termination with uncertainty estimation (spend more time on the portions of the tree that are more volatile).