What the heck

[saratoga]

Hi,
Am long time user of touring plans. Just did a plan for this Saturday at MK and the lightning lane suggestions were for Pirates and later Small world. I asked myself “what the heck”. I ran a different one a couple of days ago and it said to use Mickeys Philharmagic. Can someone explain the logic of these Touring Plan generated LL recomendations. I am using the canned 1 day adult plan.

Thanks

[Jeff_AZ]

My personal opinion is that Genie+ is too complex for any simulation to properly account for, at least one that can be run quickly and conveniently for many users.

So what I recommend to do is to hard code your own Genie+ picks. Basically, put in a plan assuming no Genie+ and Optimize.

Then, at 7am on the day of, get a LL for the first ride of your day that has a long-ish wait (>30 minutes). Plug that into your plan and re-optimize. Then when you are eligible for another LL, get one for the next available ride in your plan that has a long wait and optimize again. etc.

You can also do this exercise in advance using the “manually enter your LL return times” function. In that case, you should “Evaluate,” NOT “Optimize.”

[willsob]

Yeah, you definitely can’t blindly follow the suggestions. Plan out your start but be flexible.

With the current G+ system I’m basically looking at a TP as a very loose outline or a worst case scenario. If everything I want to do fits in the number of hours available, then I can use G+ to make the day better. So a good portion of the time the TP says I’ll be standing in line can instead go to shows/characters/snacks/shopping/etc.

[bebe80]

Just sharing this relevant info here.

@len

FWIW, we’re putting into production this week a new way of forecasting same-day (i.e., intraday) Genie+ return times.

We’ve been testing this new method since early December, pulling data from production every 15 minutes, every day, and forecasting to our staging environment. It is substantially better that the previous method for a bit more than half of WDW’s attractions. It’s about the same for most of the rest. The attractions for which it’s not as good as the current method are so few, by such a small amount, and for attractions that break down a lot anyway. So replacing the old method with the new is a worthwhile trade.

Days-in-advance forecasts (i.e., not same-day but for future days) are super important, too. They’re useful for crafting your touring plans and G+ acquisition strategy.

Taking inspiration from @Joshua9000’s Monte Carlo (MC) simulation method, we’ve developed a new framework for testing that approach against what we have now. As you can imagine, there’s a ton of work to do just to compare two methods, let alone develop a second one. So we’re still early there.

I’d expect to know in about a month the crowd levels and attractions for which MC is better than what we have. In the meantime, let me know if there are any questions.

Len

[len]

Hey @saratoga! Thanks for using the site all these years!

If this is your MK plan titled “012024 Magic Kingdom”, it looks like it’s suggested G+ for Haunted Mansion, Pirates, and Jungle Cruise. Did you pick those, or are they the result of optimization?

I ask because the next-longest wait after that is 20 minutes, for Peter Pan. And your average wait across all attractions for all day is projected at around 8 minutes.

Also, are you planning on getting an ILL for TRON? That might not be a bad idea, if it’s in the budget.

[jflafond_869328]

I was just about to post an identical thread !!! In my case, the TPs suggest using G+ to meet Ariel and do the Winnie the Pooh ride while waiting SB for the mountains…Seemed awfully weird to me…