Data Science Tool

Ship drop probability, pity mechanics, and expected value in one interactive view.

I built this as a fan-made simulator because loot-box discussions often stay vague right up until someone wants to know the actual odds of walking away empty-handed. This page turns that uncertainty into an exact, step-by-step probability model rather than a hand-wave or a rough binomial shortcut.

Probability modeling Decision support Scenario design

Read the case study Official game site

Unofficial fan-made tool. World of Warships is referenced here only as the subject of the analysis; this simulator is not endorsed by or affiliated with Wargaming.

Scenario Controls

Set the assumptions you want to test.

Use a preset if you want a fast starting point, then tune the numbers manually.

Scenario presets

These are illustrative setups for reasoning about odds and spend.

Base drop chance

3.5%

Per container, before pity applies.

Target containers opened

The summary tiles update for this specific spend horizon.

Chart range

150

Controls how far the curve extends for the full view.

Enable pity mechanic

Guarantee rules change the shape of the whole distribution.

Pity threshold

If no ship appears in the last N pulls, pull N is guaranteed to hit.

Chance of at least one drop --

Chance of exactly one drop --

Chance of zero drops --

Expected number of drops --

Output

Read the trade-off, not just the headline odds.

The marked point shows the current target-container setting.

How To Read It

Use the chart like a budget conversation.

At least one drop is the best answer to "what are my chances of success?"
Zero drops is the risk line that usually matters most emotionally and financially.
Expected drops helps compare scenarios, especially when a pity mechanic narrows the downside.

Methodology

Why this is not a simple binomial calculator.

Once a pity mechanic is introduced, each pull depends on the dry streak that came before it. The simulator tracks those states explicitly, which keeps the probabilities exact and makes the expected-value curve honest.

For a related portfolio case study on evaluation and decision quality, see Beyond WAPE - Building a Forecast Quality Framework.