For the last year or so I've noticed a flaw in letter allocation.
Every time I get an I the odds on a second I following it (presumably about one in 12-15?) are about one in two.
I've posted on this in chat forum and indeed the last two examples are today.
Can you revisit the algorithm for letter allocation to see its not stuck or somehow flawed.

I realize I may be looking at my sample with a biased eye but the numbers do not lie.

That certainly does not happen with my letter allocation. I have a massive document listing all my letters, and the words I played, and there is no duplication of I in sight. If there is a letter that tends to come to me in pairs, it is E. Since the letter allocation is spread across all players, not to an individual, it is probably just the luck of timing.

How is there a flaw, garry? Are you presuming the letter allocation based on Scrbgle? This game is not Scrabble. It accepts words that Scrabble does not. It doesn't accept words Scrabble does.

Are you saying there is a flaw based on what you know of the words in the English language? The game is distributing letters throughout the hour. It might be distributing the letters exactly how often a letter appears in the Oxford English Dictionary. You are merely playing whenever I is due to be distributed.

I hate to be the one to say it but *sigh* maybe the Word Wizard has I's for you.

Hay, somebody has to do it, Kyle....