Right-click your Alert Window, select Backtest Strategy, and OddsMaker runs your scan against roughly the last 64 trading days, then hands back a profit factor, win rate, and equity curve in seconds. That’s the whole mechanic. The skill is in configuring the four tabs honestly and reading the results without lying to yourself, and that’s what this guide covers: every setting, a worked example with real numbers, and the traps that make a pretty backtest fall apart at 9:31 on a live Monday.
This is for traders who already have a Trade Ideas scan and want to know whether it actually makes money before risking a dollar on it. If you don’t own the software yet, start with our Trade Ideas review to decide whether it earns a slot in your stack at all.
What you need before you start
OddsMaker is a Premium feature. The official pricing page, verified in June 2026, lists System Design & Backtesting under Premium only; the Basic tier gets you the scanner, charts, and paper trading, but no backtester. Premium runs $254 a month billed monthly or $178 a month billed annually ($2,136 a year). The full tier math, including the annual-vs-monthly decision, lives in our Trade Ideas pricing breakdown, and if you want to try Premium before committing, the free trial page covers the cheap and free routes in.
The desktop software is built for Windows; the pricing page points Mac users to virtualization tools like Parallels or a cloud Windows instance. You’ll also need a scan worth testing: an Alert Window with at least one alert and the filters that define your setup. If you’re still building that, our Trade Ideas walkthrough gets you to a working scan first.
How the backtester decides what counts as a trade
OddsMaker is event-based. Your alert is the entry signal: every time a stock would have fired your scan during the test window, the backtester enters a simulated position and exits it by your rules. That’s different from charting-platform backtests that walk one symbol through time. Here, the test sweeps every stock that matched your scan, which is exactly how you’d trade a scanner in real life.
The engine follows fixed rules, all documented in the official OddsMaker chapter of the Trade Ideas user guide:
- It works chronologically, earliest alerts first, across roughly 64 trading days of history. That’s the ceiling; you can test as little as one day.
- The current trading day is excluded. Yesterday is the freshest data you’ll get.
- It tests regular market hours only. No premarket, no postmarket.
- Once it takes a position in a symbol, that symbol is done for the day. Each day starts fresh.
- It logs the first 100 trades per day, which works out to 6,200 trades if you run the full available history.
- Simulated fills come from 1-minute open-high-low-close candles, not tick data, and the spread isn’t factored in. The documentation is blunt that the backtester isn’t designed for high-frequency or rapid scalping strategies, and that results will differ from live trading.
Hold those rules in your head while you configure. Half of all confusing backtest output traces back to one of them.
The walkthrough, tab by tab
Step 1: open the configuration window
Right-click inside your Alert Window and select Backtest Strategy. The Oddsmaker Configuration menu opens with four tabs. You can also backtest a saved strategy from the Multi-Strategy Window: right-click, select Strategies, pick the one you want, then Backtest Strategy.
Step 2: set the entry window (Entry tab)
Set a daily Start Time and End Time for new entries. Leave it at 9:30 am to 4 pm and the strategy hunts entries all session. Set the start to 11:00 am and nothing fires before then. This is the single most useful tab for day traders, because most setups have a personality by time of day: an opening-drive scan that prints money from 9:30 to 10:30 is often a chop fest by lunch.
Step 3: define the time-based exit (Timed Exit tab)
Three options. Minutes After Entry closes the trade after a set hold time; enter less than that window before the close and the trade exits at the closing price instead. Time of Day exits at a fixed clock time. Or exit relative to the market open or close a set number of days out, which is the swing-trader option for multi-day holds.
Step 4: add a target and stop (Risk Management tab)
Set a Profit Target and Stop Loss per share, in dollars or percent. Pick one basis and check its box before filling in numbers.
There’s also a Wiggle checkbox that widens your stop by the stock’s 15-minute volatility multiplied by its relative volume, so the stop loosens when the stock runs hot. Two cautions, both straight from the documentation: the Wiggle is not capped, so it can stretch your risk well past what you intended, and Brokerage Plus, the live-trading module, doesn’t have a Wiggle at all (it offers a defined offset instead). A backtest tuned around the Wiggle is testing exits you can’t replicate live.
Targets and stops can also be based on a Trade Ideas filter rather than a static number. Powerful for research, but the user guide spells out the catch: in live trading, orders go straight to your broker, and brokers need static stop and target prices. The backtester can test exits the real world won’t execute. Treat filter-based exits as analysis, not a plan.
Step 5: trailing stops and test length (Advanced tab)
The default exit condition here is None, which is fine for a first pass. You can add a trailing stop based on a percentage or on 15-minute bars, or use a second Alert Window as the exit signal. Same live-trading caveat applies: the only moving stop Brokerage Plus supports is a traditional trailing stop.
Two settings on this tab matter for everyone. Use Columns from Strategy limits the results to the filters already in your scan; Use All Available Columns lets the optimization tab dig through every Trade Ideas filter afterward, at the cost of a heavier run. And Test Days sets the history length, from 1 to 64 days.
Step 6: run it and read the Summary tab
Hit run and the results window opens with your strategy’s alerts and filters iconized in the top corner (hover any icon for its name and value). The Summary tab is the verdict screen:
- Profit factor: the sum of all profits divided by the sum of all losses. Above 1 the strategy made money; the user guide notes most traders aim for 2 or better.
- Win rate, trade count, average winner vs average loser, strategy return, and projected annual return, all shaded green or red by magnitude.
- Buying power: the maximum capital the strategy demanded. Set your Starting Equity to your real account size; if required buying power exceeds four times that figure, the field turns red, because 4:1 is the intraday leverage most brokers allow. A red field means you couldn’t have taken every signal, so the equity curve is partly fiction for your account.
- Position size and commissions: size each trade in dollars or shares, and enter your broker’s per-trade and per-share commission. The equity curve then draws two lines, darker green for gross and lighter green for net after commissions. The gap between them is what your broker eats.
The small calendar bottom-left color-codes every test day; hover for that day’s win rate, profit, and average gain, and double-click any day to open the Trades Window with every fill: symbol, alert type, entry and exit time and price, share size, gross profit. Below the calendar you’ll find the longest winning and losing streaks and the biggest single days, which tell you what the drawdowns will feel like. The Daily, Drawdown, and Buying Power graphs round out the picture; the documentation’s own standard for a good equity curve is a smooth 45-degree climb without deep dips.
Step 7: optimize (Optimization tab)
This is where OddsMaker beats eyeballing. The Optimization tab breaks your results into segments by Price, Time of Day, and Symbol (each gets its own sub-tab), plus every other filter under Other Filters, and shows profit factor, win rate, average gain, total gain, and trade count per segment. The graph on the right plots whichever metric you pick, with a red equilibrium line (drawn at 1 for profit factor) so the losing segments jump out.
Adjust the Price Interval or Minute Interval fields to widen or narrow the segments, and use the Having at Least field to hide segments with too few trades to mean anything. Double-click a segment to see its individual trades.
The workflow the user guide itself recommends: duplicate your Alert Window, tighten the weak filter, rerun the backtest on the copy, and compare. Repeat until the curve stops improving, then stop. More on why you should stop below.
Step 8: save your work (Settings tab)
The Settings tab shows the configuration of the run and the housekeeping buttons. Save Detailed Results exports every trade to a CSV for Excel or your own analysis; note it saves trade data only, so save the Alert Window itself to the cloud to keep the configuration. Load Results reimports an edited CSV, Run Again reopens the configuration window, and Collaborate shares the strategy with other users.
A worked example: testing an opening-drive long
Here’s a full hypothetical pass so the numbers have somewhere to live. Say you run a momentum scan and configure: entries 9:30 to 10:30 only, exit 30 minutes after entry, profit target $0.30 per share, stop loss $0.15, no Wiggle, no advanced exit, 40 test days, 100 shares per trade, $0.005 per share commission, $30,000 starting equity.
The run comes back with 240 trades: 125 winners averaging $0.31 per share, 115 losers averaging $0.19. Notice the average loser is bigger than the $0.15 stop you asked for. That’s not a bug; the documentation warns that with tight stops, the estimated exit can land well past your level, which mirrors what slippage does to a real stop on a fast tape.
The profit factor is (125 × $0.31) / (115 × $0.19) = $38.75 / $21.85 = 1.77. Net edge is $16.90 per share across 240 trades, or about $0.07 per share per trade: $7.04 per 100-share trade gross. Your $1.00 round-trip commission knocks that to $6.04, a 14% haircut you’ll see plainly as the gap between the gross and net equity curves. Over the test, that’s roughly $1,450 net on $30,000.
A 1.77 profit factor is workable, not finished. So you open the optimization tab, and the Time of Day segments show entries after 10:00 running below a 1.0 profit factor while the 9:30 to 10:00 block sits near 2.4. You duplicate the window, cut the entry window to 9:30 to 10:00, and rerun. Fewer trades, better curve. That’s the loop.
Optimizing without fooling yourself
The optimization tab is a curve-fitting machine if you let it be. Stack enough filters and any 64-day window will look brilliant, because you’ve stopped describing an edge and started memorizing the past. Three defenses:
- Watch the trade count. Every filter you add shrinks the sample, and the Having at Least field exists precisely because thin segments aren’t evidence. A 3.0 profit factor on 14 trades is a coin flip wearing a suit.
- Remember the window. Sixty-four days is one market regime. A strategy tuned on a quiet, grinding tape has never met a volatile one.
- Follow the sequence the documentation itself prescribes: backtest and optimize first, then paper trade the strategy under real market conditions, and only then go live. Paper trading is where the spread, the fills, and your own hesitation get their vote.
A passed backtest is a hypothesis with supporting evidence, nothing more. Most day traders lose money, including plenty who backtested first, and the gap is usually execution and discipline rather than the scan.
What the OddsMaker can’t tell you
Every limit below is stated in the official documentation; each one has a practical consequence.
It can’t tell you your real fill. Simulated prices come from 1-minute candles with no spread, so on thin, low-float names the backtest is systematically optimistic, and the tool is explicitly not built for rapid scalping. It can’t see further back than about 64 trading days, so you’re testing one regime, not a career. It can’t see today; the current session is always excluded. It can’t test premarket setups, since only regular hours are simulated, even though the scanner itself runs premarket live. And on a busy scan it caps at 100 trades a day, so a broad strategy may show fewer trades than your Alert Window actually fired.
None of these kills the tool. They define what a pass means: promising on recent, regular-hours, liquid-enough conditions, pending a paper-trading audition.
Where to go next
The backtester is only as good as the scan you feed it, so the natural next step is our guide to the best Trade Ideas scan settings to give it sharper raw material. When a strategy survives both the backtest and the paper-trading audition, log every live trade against the backtested expectancy in our free trading journal template; the journal is where you find out whether the edge survived contact with you. And if you’ve been reading along without a subscription, the full Trade Ideas review covers whether Premium’s price tag makes sense for how you trade.
FAQ
Is OddsMaker included in Trade Ideas Basic?
No. Backtesting is listed as a Premium-only feature on the official pricing page as of June 2026. Basic includes the scanner, charts, and paper trading; Premium adds the backtester, Holly AI, and automation. See our pricing breakdown for the full tier comparison.
Why don’t my backtest results match live trading?
The simulation prices trades from 1-minute candle data, doesn’t factor in the bid-ask spread, and estimates stop-loss exits from historical data, so tight stops in particular often show larger losses than requested and live fills will differ. The official guidance is to paper trade a backtested strategy under real conditions before trading it with money.
How far back can a backtest go?
Up to roughly 64 trading days, the full depth of the database. The minimum is one day, and the current trading day is never included.
Can I backtest a premarket strategy?
No. The simulation covers regular market hours only, so entries begin at 9:30 am Eastern even if your scan fires premarket alerts live.
Why does the backtest show fewer trades than my scanner fires?
Three rules thin the count: one position per symbol per day, a cap of 100 trades per day, and your Entry tab time window. A scan that re-alerts the same gapper twelve times still produces one simulated trade that day.
How do I export the results?
Open the Settings tab in the results window and select Save Detailed Results to get a CSV of every trade. The file contains trade data only, so save your Alert Window to the cloud to preserve the configuration, and use Load Results to bring an edited CSV back in.