Tick, volume and range charts differ from time charts in one thing: what closes a bar. A time chart closes every bar on the clock, while a tick chart closes after N trades, a volume chart after N contracts and a range chart after price covers a fixed distance, so activity-based charts print many bars when the market is busy and few when it is quiet. All of them draw OHLC bars (open, high, low and close) from the same stream of exchange trades.
Time charts cut the trade stream by the clock, while tick, volume, range and Renko charts cut it by trading activity or price movement, so they print more bars when trading is busy and fewer when it is slow. No chart type holds more information than the trades underneath it, and the tick chart is the least portable: the same N-tick setting prints different bars on data feeds that count trades differently.
What does each chart type actually sample?
Time charts sample the clock, tick charts the number of trades, volume charts the contracts traded, and range and Renko charts the distance price travels.
- Time bars: a 5-minute chart prints 12 bars an hour and a 1-minute chart 60, whether the hour trades 1,000 contracts or 200,000.
- Tick bars close every N trades. Sierra Chart's chart settings documentation is blunt: "A tick is a trade."
- Volume bars close when the contracts traded reach N, however many trades that takes.
- Range bars close when the bar's high-low span reaches a fixed range, counted in ticks of price (the other meaning of "tick": the minimum price increment, 0.25 index points on ES). The same page gives the arithmetic: "if the Tick Size is .25 and you have a range of 8 Ticks, you will see 2.00 Range displayed."
- Renko bricks print only when price moves a fixed brick size beyond the previous brick.
Figure 1 cuts one illustrative price path four ways (5-minute, 2,000-tick, 5,000-contract volume and 2-point range bars): the time chart spaces its cuts evenly along the clock, and the other three crowd theirs into the bursts of trading.
How many bars does each chart type print per hour?
A time chart prints the same number of bars every hour, while tick, volume and range charts print most of their bars in the busy open and close and few over lunch. Intraday volume clusters at the open and close, a pattern Admati and Pfleiderer modelled in The Review of Financial Studies (1988), and the open, lunch and close rhythm of the ES session is its everyday form.
The table applies each chart's closing rule to one ES session. The inputs (trades, contracts and directional price travel per hour) are illustrative, shaped like the usual busy open, quiet lunch and busy close, and are not measured CME statistics. Formulas: time bars per hour = 60 / bar minutes; tick bars = trades / N; volume bars = contracts / V; range bars = directional travel / range size (an approximation, because reversals and gap handling change exact counts); minutes per bar = 60 / bars.
| Hour (ET) | Trades | Contracts | Travel | 5-minute bars | 2,000-tick bars | 5,000-contract bars | 2-point range bars |
|---|---|---|---|---|---|---|---|
| Open 09:30-10:30 | 60,000 | 200,000 | 40 pts | 12 (5.0 min each) | 30 (2.0 min) | 40 (1.5 min) | 20 (3.0 min) |
| Lunch 12:00-13:00 | 12,000 | 40,000 | 8 pts | 12 (5.0 min) | 6 (10.0 min) | 8 (7.5 min) | 4 (15.0 min) |
| Close 15:00-16:00 | 40,000 | 150,000 | 24 pts | 12 (5.0 min) | 20 (3.0 min) | 30 (2.0 min) | 12 (5.0 min) |
| Three hours | 112,000 | 390,000 | 72 pts | 36 | 56 | 78 | 36 |
| Share of bars in the open hour | 33.3% | 53.6% | 51.3% | 55.6% | |||
| Open-hour bars / lunch-hour bars | 5.0x | 5.0x | 5.0x | 1.0x | 5.0x | 5.0x | 5.0x |
The 2,000-tick chart prints a bar every 2.0 minutes in the open hour and every 10.0 minutes at lunch, while the 5-minute chart holds at 12 bars. Each activity chart puts more than half of its three-hour total into the open hour; the time chart puts a third there. The three activity charts share a 5.0x open-to-lunch ratio only because every illustrative input falls fivefold at lunch. At the close, where the average trade is larger (3.75 contracts against 3.33 at the open), the volume chart keeps 75% of its open-hour pace and the tick chart 67%, so tick and volume bars diverge whenever trade size changes. Figure 2 plots bars per hour for the four settings across the three hours.
Are tick and volume bars statistically better than time bars?
For modelling returns, the research says yes: price changes measured over a fixed number of trades come much closer to a normal (bell-curve) distribution than price changes over a fixed stretch of clock time. Mandelbrot and Taylor stated the idea in Operations Research in 1967: "Price changes over a fixed number of transactions may have a Gaussian distribution. Price changes over a fixed time period may follow a stable Paretian distribution, whose variance is infinite." A stable Paretian distribution is fat-tailed: extreme moves turn up far more often than a bell curve allows.
Clark (1973, Econometrica) modelled prices as running on a clock set by volume. Ané and Geman tested both clocks in The Journal of Finance (2000) and concluded that "the cumulative number of trades is a better stochastic clock than the volume for generating virtually perfect normality in returns" (a stochastic clock advances with market activity, not seconds). López de Prado's Advances in Financial Machine Learning (Wiley, 2018) recommends tick, volume and dollar bars over time bars for quantitative research.
Research on trade clocks concerns the statistical distribution of returns, which risk models and backtest statistics depend on. We found no study showing that a discretionary trader makes more money looking at a tick chart. Every bar type summarises the same prints and loses detail; the full record is what footprint charts and cumulative delta read directly.
Why does a 2,000-tick chart look different on another platform?
A 2,000-tick chart looks different on another platform because a tick count depends on how the data feed counts trades, and feeds do not all count the same exchange event the same way. On CME Globex, according to the exchange's MDP 3.0 Trade Summary documentation, "A Trade Summary message represents a distinct match comprised of all orders that traded together as the result of a single aggressing order, elected stop order, mass quote, or a market state event." The same message can also carry order-level detail: the ID and fill quantity of each order in the match.
A data feed can report one CME match as a single trade or as several sub-trades, one per filled resting order. Take one illustrative match: an aggressive buy for 12 contracts (the aggressor, the order that crosses the spread; see order flow and the DOM) fills at one price against five resting sell orders of 5, 3, 2, 1 and 1 contracts. Feed A reports the summary: 1 tick, 12 contracts. Feed B reports each fill: 5 ticks, still 12 contracts. Price and volume match; the tick count differs fivefold. Figure 3 traces the match from the aggressor through the five resting orders to the two feeds.
Sierra Chart's help page on inconsistent bar charts states the consequence: "Number of Trades, Volume and Range bars are very sensitive to the underlying data you are using." Two Number of Trades charts (Sierra's name for tick charts) with identical settings "probably will not exactly look the same" when the services behind them report trades differently or one of them skips trades (see market data conflation and throttling), and the page's instructions include "Set the Combine Trades into Original Summary Trade setting to No." The setting, listed under Trades Combining Method in Sierra Chart's chart settings, does what its name says and merges sub-trades back into the exchange's original summary trade. One platform can therefore count the same match either way, and the same feed differences explain why a chart price can differ from the broker's.
Tick chart vs volume chart: which clock holds up across feeds?
Volume bars hold up when feeds bundle trades differently and tick bars do not, because a match reported as one trade or as five sub-trades carries the same contracts at the same prices. A volume bar's boundary can still shift by the one trade that crosses its threshold, and a feed that drops trades distorts both clocks. Scaling the five-fill example to the illustrative session, with Feed B splitting each summary trade into 2.5 sub-trades on average (an illustrative ratio, not a measured one):
| Hour (ET) | Feed A ticks | Feed A 2,000-tick bars | Feed B ticks | Feed B 2,000-tick bars | 5,000-contract bars (both feeds) |
|---|---|---|---|---|---|
| Open 09:30-10:30 | 60,000 | 30 (2.0 min) | 150,000 | 75 (0.8 min) | 40 |
| Lunch 12:00-13:00 | 12,000 | 6 (10.0 min) | 30,000 | 15 (4.0 min) | 8 |
| Close 15:00-16:00 | 40,000 | 20 (3.0 min) | 100,000 | 50 (1.2 min) | 30 |
| Three hours | 112,000 | 56 | 280,000 | 140 | 78 |
In the open hour an average Feed A tick bar holds about 6,667 contracts and a Feed B tick bar about 2,667, so one setting describes bars of very different weight. To approximate Feed A's chart, Feed B needs 5,000-tick bars (2,000 x 2.5). Figure 4 sets the two feeds' 2,000-tick bars per hour side by side, with the unchanged volume bars as a reference.
The same 2,000-tick setting printed 56 bars on one illustrative feed and 140 on the other, from identical trading.
Statistically, research favours the trade count: Jones, Kaul and Lipson found in The Review of Financial Studies (1994) that "it is the occurrence of transactions per se, and not their size, that generates volatility". Portability favours volume: a trader who compares charts across platforms or shares settings gets a consistent clock from volume bars, while a trader who stays on one feed can use tick bars without that cost.
Range bars vs Renko: do range bars really have no gaps?
Not in the market: range bars show no gaps only when the platform builds them to hide gaps, and the prices underneath still gap. Sierra Chart offers several variants, including "Range Per Bar - Fill Gaps", which fills price gaps between bars, and "Range Per Bar - True", which leaves the open, high, low and close unadjusted. Sierra Chart's documentation says that for standard range bars built without tick-by-tick data, an oversized bar is split and "The open of the next bar will always be moved one tick outside of the prior bars range."
Range bars also hide time. In the illustrative session a 2-point range bar takes 3.0 minutes on average in the open hour, 15.0 minutes at lunch and 5.0 minutes at the close, yet every bar is drawn the same size. Figure 5 places those equal-looking bars on a real time axis.
Renko goes further: a brick prints only when price moves a full brick beyond the previous brick, so back-and-forth trading inside a brick leaves no mark. Against tick bars, the difference shows in heavy two-way trading that goes nowhere: a tick chart keeps printing bars while range and Renko charts wait for price to move.
Which chart type is best for scalping, swing trading and backtesting?
No chart type is best in general; the right one depends on the holding period, the hours traded and the data a strategy was tested on.
Scalping. Tick and volume bars give finer bars when the market moves fast and stop printing near-empty bars at lunch. The best tick chart for ES, NQ or MNQ is not a universal number: choose the bars per hour you want in the hour you trade, then divide that hour's trade count on your own feed by it. On Feed A's illustrative open, 30 bars an hour means 60,000 / 30 = 2,000 ticks; on Feed B the same target means 5,000.
Indicators. A lookback counted in bars covers different spans of time on activity charts: a 20-bar moving average on the illustrative 2,000-tick chart spans about 40 minutes at the open-hour pace and 200 minutes at the lunch pace.
Swing trading. Time bars line up with the events that move positions held for hours or days: the open, scheduled data releases and daily settlement.
Backtesting. Activity bars can be rebuilt exactly only from trade-by-trade history, which is why tick data beats bar data for backtests of these charts. The same Sierra Chart help page adds a trap: "The default Intraday Data Storage Time Unit that Sierra Chart uses when collecting data in real-time and downloading historical Intraday data, is 1 second." Tick, volume or range bars built from 1-second records cannot close inside a second, so they can differ from bars built trade by trade. Platform defaults differ; see Sierra Chart for prop trading and NinjaTrader vs Quantower charting.
Where does a trade copier fit, and where is it not the answer?
A trade copier removes one feed problem: when the decision is made once, on one master chart, every follower account receives the same orders even if its own feed would draw a different tick chart. Thor, this blog's own product, copies orders from a master account to follower accounts server-side. A copier does not validate a backtest, reconcile feed differences inside your own analysis or choose a chart type. A trader with one account, or whose charts and backtests already share one feed, gains nothing here from a copier.
Go deeper
- Tick Data vs Bar Data: What "Historical Data" Actually Means When You Backtest
- Why Your Feed Shows Fewer Ticks Than Actually Traded: Conflation and Throttling
- Footprint Charts and Cumulative Delta Explained
- NinjaTrader vs Quantower: The Best Charting Platform for Futures
Frequently asked questions
Why do traders use tick counts like 233, 610 or 1,597?
Those are Fibonacci numbers, and choosing them is a convention, not a finding. We found no research showing that a Fibonacci tick count performs better than any other count.
Do tick charts repaint?
Not while the chart stays on one live feed: completed tick bars do not change and only the forming bar updates. They can redraw after a reload if the platform rebuilds history from stored data with a different granularity or trade grouping than the live stream used.
Why does my 5,000-contract volume bar show more than 5,000 contracts?
Because on some platforms the trade that crosses the threshold is not split between bars. Sierra Chart's chart settings documentation says each volume bar "will be equal to or greater than" the setting, so a 7-lot arriving at 4,998 contracts closes the bar at 5,005.
Is a seconds chart a time chart?
Yes. A 30-second chart closes a bar every 30 seconds regardless of activity, so it prints 120 bars every hour, at the open and at lunch alike. Shorter intervals add resolution but keep the time chart's mismatch between bar count and activity.
Is Heikin-Ashi a different bar type?
No. Heikin-Ashi changes how bar prices are calculated, not when bars close. Its close is the average of the bar's open, high, low and close, and its open is the midpoint of the previous Heikin-Ashi bar's open and close, so it can sit on time, tick, volume or range bars and inherits their sampling.
Does the chart type change my fills or slippage?
No. Futures orders fill at the exchange against the order book, whatever chart produced the signal. Chart type changes when a signal fires, and so the market you meet, but not how the exchange fills an order once it arrives.
Do tick charts work in the overnight session?
Yes, but they slow down because far fewer trades print overnight. A tick count calibrated on the cash open prints much longer bars in the overnight session, so each session needs its own setting.
Do tick charts work for forex and CFDs?
Yes, but a tick there usually means a price update, not a trade. Spot forex and CFDs trade over the counter with no central record of trades, so platforms such as MetaTrader count price updates from the broker's feed as ticks, and the volume shown is tick volume, a count of those updates. A tick chart on a CFD therefore measures one broker's quote activity, not market-wide trading.
Sources
- Sierra Chart (current), Chart Settings documentation
- Mandelbrot, B. and Taylor, H. M. (1967), On the Distribution of Stock Price Differences, Operations Research 15(6)
- Ané, T. and Geman, H. (2000), Order Flow, Transaction Clock, and Normality of Asset Returns, The Journal of Finance 55(5)
- CME Group (current), MDP 3.0 - Trade Summary, CME Group Client Systems Wiki
- Sierra Chart (current), Number Of Trades, Volume or Range Bar Charts Do Not Match Setting or Are Not Accurate or Are Inconsistent (help topic 47)
- Jones, C. M., Kaul, G. and Lipson, M. L. (1994), Transactions, Volume, and Volatility, The Review of Financial Studies 7(4)