Reading a real-money record: the Autotrader

A pick record is graded at the published price. A real-money record is graded at the price actually paid, after fees, at the size actually filled. The gap between them is the cost of execution: prices that moved before the order filled, fees, and orders that did not fill at all.

Paper records skip those costs. A paper fill assumes the price shown was really available at the size wanted, so a paper record is an upper bound on what real money would have earned.

Read net profit and loss in dollars, after fees, and the number of settled trades behind it. Then ROI, which divides that net result by the money staked. A win-loss record alone says little, because trades on prediction markets are bought at very different prices.

Read the start date. A record that begins at a chosen moment can leave out an earlier stretch. A published start date, and the reason for it, let you judge that choice.

Kalshi is a US prediction-market exchange regulated by the CFTC. You buy yes or no contracts from other traders rather than betting against a book. A contract settles at $1 if its outcome happens and $0 if not, so its price in cents reads as a probability.

Two things follow. Fees come out on both the buy and any sale, so they weigh more on a small edge than sportsbook juice does. And positions can be sold before the game ends, so a trade can close at a profit, at a loss or at settlement.

The MVP Capping Autotrader trades picks from selected MVP Capping pick models with real money on Kalshi, and its record is public at /autotrader. Profit and loss there is net of Kalshi's fees on both the buy and the sale, and entry and exit prices are the prices actually traded. The trade log shows how each trade closed: settled, take-profit, stop-loss, pre-game exit, or sold by hand.

The record page states when the record starts. Open positions on games that have started are shown publicly; positions on games not yet started are visible to Pro members. The public record is a file rewritten every five minutes, so it can trail the exchange by a few minutes.

Alongside the live trader, paper models run the same process on the same live prices without placing orders. Their records are useful for comparing ideas, but they do not pay fees on failed fills or face prices that move while an order waits. Expect a paper record to look better than the same process would with real money.

When a service shows only paper results, ask why. When it shows both, compare them. A large and growing gap between paper and live is the cost of execution, and it is the number a real-money bettor actually pays.

The Autotrader stakes a fixed share of the account's cash balance on each order. When the balance grows, stakes grow; when it shrinks, they shrink. That means the dollar record weights later trades by the balance at the time, and deposits or withdrawals change stake sizes too.

Read ROI as well as dollars for that reason. ROI divides by the money staked, so it compares periods with different stake sizes on the same scale.

The Autotrader's results are not the results you would get following the same picks at a sportsbook. Prices, fees, timing and stake sizing all differ. Read it as evidence of what one real-money process did, win or lose, and read the how-it-works page for the rules it trades by.