There isn't much I can do about the reason statement when I'm only allowing one positive balance. We can guarantee that no balances would be worse. 
For giggles, I might pre-pay the utility bill just to see what happens on EX and TU. I wouldn't be surprised to find that the effect is nothing if I let my normal one to two cards report balances.
@Anonymous wrote:Agreed that the 15-20 point drop associated with "no revolving credit use" is > the 1-2 (say) point drop associated with "too many accounts with balances."
I agree that would be a good test for S&G's on your profile... going to AZEO on revolvers and trying to get that utility bill to report $0 one month to see what happens.
The score drop associated with too many accounts with balances can be rather substantial with mortgage Fico scores. Of course, as more accounts report, point drop associated with the "too many accounts with balances" reason code increases. Depending on profile, dropping from 2 reporting to 1 reporting may gain up to 10 points - IMO (all else being equal). I suspect this is the primary factor limiting my EQ, TU and EX Classic Fico 04 scores (minimum 3 accounts with balances, 2 cards) - No more scoreable inquiries as of this month .
@Anonymous wrote:From what I understand, it's only $30.
I think you can find some similarities across the board though, which can be helpful to some in being able to guesstimate scores. For example, if 8 out of 10 people say that their Score X runs 10-15 points higher than their Score Y, that's pretty strong data suggesting that Score X generally on most profiles runs higher than Score Y. Then you can look to the 2 out of 10 people who state that their Score X isn't higher than Score Y and perhaps identify the profile similarity between those two individuals that possibly isn't present or oa factor across the 8.
For that sort of statistical analysis, there's the standard p-test: you need far more datapoints for trying to do analysis like this, 10 isn't close to enough and even 100 has a non-trivial error rate.
It's statistics 
.
Again if you really want to know where you stand, just pay for the pull: otherwise I will suspect it will have little more value than estimating scores via Oujia board, and basically the same value as trying to estimate FICO from one's Vantage scores or vice versa: they're just different, even if the FICO 8 industry options are of course closer to FICO 8 than Vantage ever will be.
@Revelate wrote:For that sort of statistical analysis, there's the standard p-test: you need far more datapoints for trying to do analysis like this, 10 isn't close to enough and even 100 has a non-trivial error rate.
It's statistics
I agree with what you're saying if you're talking just scores, but many of the individuals in this thread have been kind enough to include a good chunk of the profile data that goes into those scores, making the numbers far more meaningful than just a basic statistics activity.