If your four cases cover all possible hands, then yes - you will want four values. It doesn't matter whether those cases are 0-3 or 1-4, but you would use three IF statements and have the final "else" be the fourth value. If you do that, though, you will end up including limped pots in the "else" case and I notice they're not on your list. With a grouping stat like this it isn't possible to exclude certain types of hands from the grouping altogether as they will always end up with the "else" case value. If you want to exclude certain hands altogether then you will need to use a filter as well.
I do have a limped pot stat, I just didn't add it to this. I have now added it, but are not been grouped separately
Also, this means all the hands that did not see a flop "preflop pots" will need to be included? How is this included in the "else[]" expression, or does it stay blank?
One thought just occurred to me - a "squeeze pot" will generally be a 3bet pot, so if you were to switch to using "char_length(cash_hand_summary.str_aggressors_p) = 3" to test for 3bet pots, then you would need to test for the squeezing case BEFORE that, otherwise, the squeezes will be counted as simple 3bet pots. (It looks as if the expressions you posted above explicitly exclude squeezes from the 3bet pots, so using those wouldn't have that problem.)
Also be aware that using "char_length(cash_hand_summary.str_aggressors_p) = 4" would exclude 5bet and higher pots, so you'd probably want to use "char_length(cash_hand_summary.str_aggressors_p) >= 4" in that case.
If which case would use of "char_length(cash_hand_summary.str_aggressors_p) = 3" actually simplify my squeeze stat? I would still need to use "not (cash_hand_player_statistics.flg_p_squeeze or cash_hand_player_statistics.flg_p_squeeze_def_opp)", correct?
Also, considering people have a tendancy to call 4bets on occasions (be it either a leak, tricky play or being deep stacked), would using "char_length(cash_hand_summary.str_aggressors_p) = 4" & "char_length(cash_hand_summary.str_aggressors_p) >= 5" be suitable?