summary: the IB report shows how price behaves after the first hour range. this article is the trade-it layer: the one number to read, what to actually do with it, how the retracement and level counts are counted, a worked example, how to set it up so it matches your chart, and what to check when the numbers look off. for the full breakdown of all 11 variants, see which IB report should I use?

the initial balance (IB) is the range price builds in the first 60 minutes of your session. the report answers one question: after that range is set, does price break above it, break below it, break both sides, or stay inside?

start with the standard view and read the overall breakout rate and the direction lean. that's your foundation.

rule of thumb: look for a read of 65% or higher before you lean on it, ideally 70%+. a 70/30 directional lean is something to trade with. a 52/48 split is the data telling you there's no edge today.

the breakout rate tells you whether to trade the break or the range, and the lean tells you which side.

if the IB breaks consistently (say 80%+) with a clear directional lean, the first hour range is a launch point. plan to trade the break in the direction the data favors. pull up by-levels to set your take-profit at an extension the data actually reaches, and use by-retracement if you'd rather enter on the pullback than chase the break.

if the IB holds most days (a high "no break" rate), the range itself is the setup. that's a fade-the-edges day, not a breakout day.

one more read worth knowing: by-rejection tells you which side formed first, and the side that forms first usually isn't the side that breaks. that gives you a directional lean before the break even fires.

the by-retracement view trips people up because the level counts don't add up to your total days — and they aren't meant to. the levels are cumulative: they nest inside each other.

price that retraces to the 75% level had to move through 25% and 50% on the way, so that day is counted at 25%, at 50%, and at 75%. each level is really telling you "how often price retraced at least this far" — not "how often it stopped exactly here." that's why the deeper levels show smaller counts, and why one day can show up at several levels at once.

by-levels works the same way on the extension side: reaching the 50% extension means price tagged 25% first. read both as "reached at least this far" and the numbers stop looking contradictory.

say you're trading NQ on the NY session. standard shows a strong upside breakout rate, by-rejection shows the low formed first (a bullish lean), and by-levels shows price reaches the 50% extension a high share of the time.

the plan that falls out of that: trade the upside break of the IB high, set your first target at the 50% extension level, invalidate if price closes back inside the IB. you built an entry, a target, and an exit straight from the data instead of guessing.

results take customization and reps. the data gives you the edge, your process and risk management are what turn it into a result.

the IB is defined by your session. an IB on the NY session (9:30 to 10:30 AM ET) measures something completely different than an IB on London or Asia. set the session to the one you actually trade.

then match your TradingView chart to it: set your TradingView timezone to match the session (NY = America/New_York) so your levels line up with the report. a mismatched timezone is the number one reason the numbers look wrong.

one more setting to know: the IB reports can filter by IB size — small, medium, or large first-hour ranges. with a size filter on, the report only counts days in that bucket, so the day total drops below the calendar days in your lookback. that's expected. clear the filter if you want every day back in the sample.

if the report disagrees with what you see on your chart, check these in order before assuming it's broken:

most mismatches are one of these. the full walkthrough is in why your edgeful data doesn't match TradingView, and the IB-specific cases are in why doesn't my number match the report?. if everything matches and the number still looks wrong, that's worth reporting, not trading around.

check the sample size before you trade a stat. a 90% breakout rate built on 6 occurrences isn't reliable. and a number that's held across the 1-year, 6-month, and 3-month lookbacks is far more trustworthy than one that only shows up on a single window.