Event backtesting: test an option strategy around RBI, Fed, Budget and election days
Pick the kind of event, how many sessions before and after it, and the trade. The backtest runs that trade on every matching window and reads the result event by event.
What event backtesting answers
An ordinary backtest asks how a strategy did across a stretch of dates. An event backtest asks a narrower question: how did it do around one kind of day. A short straddle held into RBI policy days, a long strangle bought before Lok Sabha results, a spread placed the morning after a US Fed decision. The events come from the same sourced list that Market History shows, so every test day can be traced to the decision it stands for.
Set up a test in four steps
- Index. NIFTY, BANKNIFTY, FINNIFTY, MIDCPNIFTY or NIFTY NEXT 50. Each choice says from when its recorded option prices start.
- Events. Event types in five groups: central banks (RBI and Fed cuts, hikes and holds, Fed balance-sheet moves, other central banks), elections (Lok Sabha results, exit polls, state and US elections), budget and policy (Union Budget, tax, SEBI and F&O rules), shocks (war, tariffs, oil, COVID, credit stress, index inclusion) and economic data (US CPI releases).
- Window and lookback. Up to five trading sessions before and after the event day, and how many years back to look. The window counts trading sessions, so the session before a Monday decision is the Friday before it.
- Strategy and rules. The legs, from a sample, a saved Builder strategy or your own, then entry and exit times, stops, targets and trading costs. The expiry either covers the event, the first expiry on or after it, or is each session's nearest, as in a normal run.
A short review states the whole test in one sentence before it runs. After it, the settings sit on one page above the results, and any step can be reopened from its summary chip.
The day that counts as the event
An event is placed on the first NSE session that traded it, which is not always the day it happened. A Fed decision lands after the Indian close, so the next morning is the event session. A decision on a Sunday meets Monday's open. When two windows overlap, a session belongs to the nearer event, so one session is never counted twice.
What you get back
- A summary row. Net P&L, win rate, events won, average P&L per event, maximum drawdown, and the best and worst event by name and date. Each figure says whether it is counted per session or per event.
- By day offset. One row for each session of the window you chose, T-2 to T+1 for example, so you can see whether the result came before the event, on the day or after it.
- By event. Every event in the test, with the index move on the event day and what the strategy made around it. Each row opens to its sessions.
- The full report. The same trade log and analytics as the backtester, folded under the event results.
The index study, before any strategy
A drawer beside the event picker shows how NIFTY 50 itself moved around the chosen events, from daily closes since 2019, with no option prices involved. It sets the event-day move beside an ordinary session's in the same years, compares it with the move India VIX priced the night before, and shows how VIX changed across the window. It answers the first question, whether these events move the market at all, before you build a trade on them.
What an event backtest cannot tell you
The strategy run uses recorded option prices, and those begin in July 2026. Most events in the list are older, so a test over several years may have only a handful of sessions it can run on. The page states that count before you run, and refuses to run on none. Modelled prices can be switched on, but they are off by default: a model has no event-day spike and no IV crush in it, which is exactly what an event test is meant to measure. Small samples are the rule here. Three RBI holds are three observations, and an average over three says very little about the fourth.
Event backtesting questions
What is event backtesting?
A backtest run only on the sessions around chosen market events, such as RBI policy days, US Fed decisions, Union Budgets or election results. You choose the event types, a window of sessions before and after each event, and a strategy, and the results are read per event and per day of the window.
Which events can I test?
Central bank decisions (RBI, US Fed and others), elections and exit polls, Union Budgets and tax changes, SEBI F&O rule changes, shocks such as wars, tariffs, oil and COVID, and US CPI releases. Each event in the list names its source.
Does it use real option prices?
Yes, by default it uses only sessions with recorded option prices, which begin in July 2026. The page shows how many of the chosen sessions are recorded before you run. Modelled prices can be included with a switch, and are off by default because a model misses the event-day behaviour being tested.
Is a good event backtest a reason to trade the next event?
No. It describes how a set of rules behaved on a small number of past events. The next event can differ from all of them, and the sample is usually small. It is a tool for studying risk, not a recommendation.
Related tools and guides
Educational analytics only: not investment advice. You are responsible for your own trading decisions.