Noblesse Oblige Grates Calculator

If you have a well-equipped team, it should be clear that opening 20 grates is not the optimal strategy to get through the sewer in Hobopolis the fastest. Rather, you can afford to sacrifice a couple of grates in the hopes that the turns saved will be greater than the number of people who take an additional turn due to missing the grates. But where is the ideal cut-off point? This simple JavaScript application will perform Monte Carlo simulation to tell just what the perfect number of grates is.

Enter the number of runners and the number of glyphs that they have below. You can also tweak some other parameters, including the assumptions on just how the glyph probabilities work. By default, the assumption is n/6 probability of passing the first test, (n-6)/7 probability of passing the second test, and (n-13)/7 probability of passing the third test, but this can be altered below.

For each number of grates, the simulation returns two numbers: the mean (expected) number of turns it will take to get through the sewers, and the standard deviation on the number of turns. So, if you're feeling risk-averse, you might pick something with a slightly higher mean but with less variance, so your risk of being screwed by the RNG is diminished.

Total number of people through sewer (max 35): Number of trials (see note below):
Calculate results for number of grates opened between and

NOTE: The calculator assumes that all players have all necessary consumables. If not, the number of tunnels will be higher than calculated here and you may be better off opening more grates.

Glyphs per player:
 1:  2:  3:  4:  5:  6:  7:
 8:  9: 10: 11: 12: 13: 14:
15: 16: 17: 18: 19: 20: 21:
22: 23: 24: 25: 26: 27: 28:
29: 30: 31: 32: 33: 34: 35:

Assume that glyphs are sufficient to always pass the first test and glyphs are sufficient to always pass the second test.

READ ME: Because this is very computation intensive and JavaScript is not a really good language for this kind of task, this script may take a while to run, depending on the number of trials chosen and the speed of your computer. You may get one or more "Unresponsive script" warnings while the simulation is running; you can either ignore them and continue or stop the script depending on how long things are taking.

With a slower machine and an older browser, 1000 trials (with 7 players) takes about 15 seconds; a newer machine and more modern browsers can do 10000 trials in the same time. More than 10000 trials is probably overkill. By default, the script will only consider between 10 and 20 grates opened, since, except for particularly strange cases, the minimum number of turns is always somewhere in this range, but you can change this if you wish. Note that increasing the total number of players increases the time required correspondingly.

The chart below lists the approximate accuracy of the returned results depending on the number of trials run.

Number of trials1005001000500010000
Accuracy on average number of turns±0.25±0.11±0.08±0.04±0.03
Accuracy on standard deviation±0.18±0.08±0.06±0.03±0.02

Click once and be patient; the simulation may take a while to run. If the button gets stuck in its disabled state, just reload the page.

Return to the main Noblesse Oblige site
Design and coding by Ragnok (#49653)