Most traders in Kalshi's 15-minute markets keep some record of their trades. Far fewer keep a record that can actually answer the question that matters: are my rules working, or am I just remembering the good rounds? This guide covers what to log on every trade, why you should also log the rounds you skip, and how to review the journal once a week without fooling yourself. It works the same whether you trade by hand or run a bot like DeskBot for Kalshi.
See DeskBot for Kalshi: $1 first month, then $19.99/mo, or $155/year
Why a journal beats memory
Short markets produce a lot of decisions. A busy morning can mean a dozen or more entries across a few series, and by the afternoon most of them have blurred together. Memory keeps the dramatic wins and the painful losses and drops the ordinary middle, which is exactly where you learn whether an edge exists. A written log turns "I think this works" into a number you can check.
What to log on every trade
Keep each entry short enough that you will actually fill it in. These fields cover almost everything you need:
- Time and market: the series and the specific 15-minute window, plus how many minutes were left when you entered.
- Side and price: Yes or No, the price you paid, and the size in contracts.
- Price you wanted vs price you got: if you placed a limit at 55c and filled at 58c, write both. The gap is your real cost of getting in.
- Fees: record the fee on each fill so your break-even math uses real numbers, not the headline price.
- Reason: one sentence on which rule triggered the trade. If you cannot name the rule, that is worth knowing too.
- Result: settled or sold early, the exit price, and the net result after fees.
Log the rounds you skip, too
This is the part most journals leave out. If you only record trades, you never find out whether your filters are helping. Say your rule is to skip any round where the order book looks stale or the price is above 90c. If you never write down the skipped rounds and how they settled, you cannot tell whether that filter saved you money or quietly cost you good trades.
A skip entry can be tiny: the market, the time, which rule said no, and how the market settled. After a few weeks you can compare the rounds you took with the rounds you passed on and see whether each filter is earning its place.
Break-even math belongs in the journal
At high prices the math is unforgiving. Buying Yes at 90c risks 90c to make 10c before fees, so you need to win more than 90% of those trades just to break even, and fees push that line higher. Add a column for the break-even win rate at your average entry price, and compare it with your actual win rate over the same set of trades. If the actual rate is not clearly above break-even over a meaningful sample, the strategy is not working yet, however good individual days felt. Our guide to Kalshi fees and break-even win rates walks through the numbers in more detail.
A simple weekly review
- Count, don't skim. Total trades, total skips, wins, losses and net result after fees for the week.
- Sort losses into two piles: trades where you followed your rules and the market simply went the other way, and trades where you broke a rule. Only the second pile is a discipline problem.
- Check each filter against the skipped rounds it blocked.
- Change one thing at a time, and note the date you changed it so later results are compared fairly.
- Write the date range on every summary, and include the losing days. A week of results is a small sample and does not predict the next week.
How DeskBot handles the log
DeskBot runs on your own Mac (Apple chip or Intel) or Windows PC and trades through your own Kalshi account using rules you set: which markets, the entry window, price caps and a daily loss pause. It keeps a log of its decisions with the reason for each trade, so the weekly review above starts from a record rather than from memory. You still need to read the log and decide whether your rules deserve to keep running. Software can follow rules consistently; it cannot make a losing set of rules profitable.
Want more reading first? Browse all of our DeskBot trading guides.
Pricing
$1 for the first month, then $19.99/month, or $155/year. It is a paid subscription, not a free trial, and it renews until you pause or cancel. See the product page for platform options.
Risk disclosure: Trading can lose money. Event contracts can settle at zero, and you can lose the full amount you put into a trade. A journal and a bot reduce mistakes but do not remove risk. DeskBot is software. It is not financial advice and does not guarantee any profit. Past results don't guarantee future results. Not affiliated with Kalshi or Polymarket. Use is subject to the DeskBot Terms of Service.