When you don't like the order they create, move things around to suit yourself, then hit the "Evaluate" button. Hitting "Optimize" means they'll have you riding Splash Mountain when lines are shortest (and the weather chilliest), rather than in mid-afternoon like you'd want to.I don't think the wait times have been overly accurate for me either. With that being said I do pay the $5 a year discounted rate to use a touring plan as I like to have an idea of what to do each day. However, I got the order we do from a mixture of The Unofficial Guide and from the fine folks here- I did not like their order.
Fwiw, I went back and looked at the "what we saw" for the dates of our trip. For higher crowd dates (the avengers weekend), what they saw was dead on with what the new predictions were. For the weekdays, they actually overestimated their predictions (at least one of those days was predicted, with the new predictions, to be a 5 and they saw a 3).As others have said, take any of the crowd predictors with a large grain of salt. I would look at as many different ones as you can find and see what the general consensus is. As others have mentioned Touring Plans recently re-did all of their predictions (for both WDW and DLR) because they were having problems. I noticed that since the start of the 60th Anniversary Celebration in DLR, they tended to be predicting crowds several levels below what they ultimately saw.
I'm curious to see what the long-term accuracy is for the new statistics. There were both under- and over-predication days prior to the update, but in general they tended to over-predict (especially in DL, the DCA predictions weren't as bad). So, it sounds like the new stats may be working okay for busy days, they're now having issues with days that are likely to be less busy. It would be pretty ironic if after revising their statistics they end up over-predicting days in general! Although, it does fit with that old saying about statistics...Fwiw, I went back and looked at the "what we saw" for the dates of our trip. For higher crowd dates (the avengers weekend), what they saw was dead on with what the new predictions were. For the weekdays, they actually overestimated their predictions (at least one of those days was predicted, with the new predictions, to be a 5 and they saw a 3).