Analysis - Shanten After Calling

Someone sent me a blog post with information on the average shanten after a player makes a call, then asked if the tonpu numbers would be different. So, I recreated the code and went into more detail. What's the average shanten after someone makes their Xth call on Y turn? What about their chance of being tenpai? Does the type of call affect it? Let's see.

You can find the original blog post here. I'm not entirely sure on their methodology, but it does say the data was from the 2016 replays. My data set is around 8 times as large, so there will be some number differences.

As for my methodology, I'm not counting closed kans, added kans, or nuki dora declarations as calls. Called kans are counted. The shanten is calculated before they discard, so -1 is a valid outcome.

Let's start with the hanchan data.

The blog's data starts at 0, but I'll start it at 1. The first five rows for average shanten are pretty much identical, which indicates that I'm likely using the same methodology. Now, we can move onto the tonpu data to see how it changes.

The numbers are pretty similar. The calls are only slightly worse than in hanchans. So, you don't really need to change how you react to calls based on the game length. Let's also check sanma for completion's sake.

These calls are much better. They're almost twice as likely to be tenpai after making their first call in the first row. The raw data for the above three charts can be found in this spreadsheet.

When gathering this data, something I wondered was whether it would change depending on whether the call was a chii or a pon. Or, going further, a kanchan chii or a ryanmen chii, or a yakuhai pon or a non-yakuhai pon. A terminal chii or a terminal pon?

Let's step back and look at chii vs pon first. For the following charts, I'll be using only the 200,000 most recent replays I have, because doing the full dataset takes eight hours and I want results.

A green background means it has the lowest tenpai chance of the options, while a red one means it has the highest. I don't know what to do with this information. Let's gather more.

Dragon pons are actually the most likely to be tenpai, which is interesting. Ryanmen chiis being most likely to be tenpai is expected. You open your hand on a good wait less often than a bad wait.

Maybe I should run those on the full dataset, but I don't think they're particularly useful, so it's probably fine like this. The raw data for those two can be found in this spreadsheet.