It’s an app.
It’s a game. Just like 2 player on the board game. Can be fast paced. I like fast scrabble. But sometimes I’d rather be lazy.
Ehh, it accommodated, so I continue to open it when I am bored.
The question. Am I playing another human? Or a computer?
'Cause come on folks, I’m really good at Scrabble(my spelling on this board would not back this up), but not a genius. And if I’m beating the computer I’m gonna go pro.
Where does one get an agent?
What I could find is that you can play either and there are bots disguised as humans. Just because you beat a bot doesn’t mean anything. It’s not like they have to have their skill level maxed out. From what I see at Reddit, they seem to average around 230 game points. Another report suggests they play to your skill level.
Are you actually interested in pro Scrabble? Cuz there are actual (in-person) tournaments at different levels. Might be a fun pastime?
I play ScrabbleGo often whilst sitting with Madame P. There are real players, and AI players, andcsome I can’t tell.
I tend to stick wirh playing real people.
I know there’s pro-Scrabble. But I’m sure I’m not that good. Just a pipe dream.
What a way to live, tho’. Right?
Anyway..on Scrabble GO! I’ve decided I don’t care who I’m playing. It’s a time waster, that requires no angst on my part because I’m a good speller. And it’s fun!
(I’m not a competitive player, myself)
I don’t think you necessarily have to “win” or aspire to be a world champion. They put you in different leagues, etc. It’s just a way to meet (and play with) other Scrabble players in an organized way.
Yeah. Not happening.
I don’t like human beans that much. Not really social.
So it is family, or unwitting visitors that I’d rather shame with my vast spelling abilities.
'Cause I’m just that petty 
I wrote an online Scrabble game where you play against the computer, but have no interest in playing it. I made it too good, and it beats the crap out of me (or anyone, I would wager). It always plays the best possible word out of every word in its dictionary, so it’s essentially impossible to beat unless you get very lucky with your draws and it gets very unlucky. If anyone’s interested in trying it, send me a PM.
Strategically speaking, is it always the case that playing the maximum-scoring word per turn would likely result in a win? There might be situations where the better long-term play would be to lure your opponents into a trap where they can’t play their tiles effectively, no? You could sacrifice a small advantage of yours to destroy a huge one of theirs, etc.?
If the algorithm is only to play the best word in a rack, that’s not the best strategy and I’d bet the best humans would consistently beat it. Time placement is important (you don’t want to leave valuable bonus squares accessible), holding back bingo stems and playing lower value words is important, when to trade in tiles, etc.
True, my program doesn’t consider any long-term strategy and just greedily makes the highest scoring move on each turn. I’d be interested to see how it plays against a very good human player.
[Moderating]
I just noticed that this is in Cafe Society. Moving to the Game Room.
Well, I found this paper:
In it, the baseline model is described as your greedy algorithm so far as I can tell. Under 2.2, it says they use a contributed and player, Sam Masing, ranked at 190 in the world, as a performance metric. This particular paper says he was “easily able to meet the baseline algorithm.”
In that paper, it looks like average Scrabble points for baseline is about 270 and the algorithm the paper was about scored 360. The top Scrabble players average over 400. The absolute elite up to 430 or so.
That said, this is just one paper. You can have your program self-play thousands of times and see what game score it gets.