The only two inputs needed for the simulation are 1) the YES price and 2) the number of contracts traded. You can make a copy of the spreadsheet and change these parameters to perform your own simulations. For simplicity, we assume zero fees and (NO price) equals exactly $1 – (YES price).
We will now go through each item in the simulator output.
Account Deltas and Open Interest Changes
The first few columns in the simulator track how much the open interest and maker/taker balances change during each trade.
Take note of a few invariants:
- For each trade type, the taker and maker always take opposite positions. One is long YES resolution, and the other is short YES resolution.
- The taker and maker YES and NO deltas always have the same absolute value. This is different from their USDC deltas, which can have different absolute values.
- Split trades always increase open interest, merge trades always decrease open interest, and swap trades always leave open interest unchanged.
Volume Metrics
There are two categories of prediction market volume metrics in common use:
- Notional Volume: number of contracts traded
- Cash Flow Volume: the amount of USD exchanged at time of trade
Computing these metrics for swap trades is straightforward. The notional volume is simply the number of contracts traded. The cash flow volume is the number of contracts traded multiplied by the share price. For both of these volume metrics, Polymarket’s OrderFilled sum gives a value that is 2x the correct value.
Computing these metrics for split trades and merge trades is more complicated because they are not swaps in the conventional sense (see the Splits and merges versus swaps section). In a conventional swap, the maker and taker experience the same amount of volume. But in a split trade or merge trade, the maker and the taker can experience different amounts of volume. We are not opinionated about whether to measure the maker’s volume, the taker’s volume, or something in between. However, simply adding the two together is double counting, just as it would be with a swap trade.
With these considerations in mind, computing the notional volume for split trades and merge trades becomes straightforward. It is simply the number of contracts that are split or merged.
Computing the cash flow volume for split trades and merge trades is the most complicated case. As mentioned above, the maker and taker will experience different amounts of cash flow. Whether to focus on the maker’s cash flow, the taker’s cash flow, the maker taker mean, or some other approximation (like share-priced volume) can depend on the context. These metrics can each produce different volume values for an individual trade. However, these metrics all produce about the same volume numbers when aggregated over longer timescales of days or months (see Figure 5 below). Thus, the exact choice of cash flow metric is often unimportant. The only important note is that summing up OrderFlow events does not agree with these volume metrics, and it instead produces a value about 2x as large due to double counting (see Figure 5 below).
