The venue gives you a line. A chart needs four numbers.

Open an event on Polymarket and the chart you get is a probability line. One number per point in time, usually the last traded implied probability, connected through the life of the market. It is a good headline. It is a bad chart.

That is not a slight. The venue is a matching engine with a reference price. A reference price is a close. Traders who grew up on TradingView then go looking for candles, because a close-only series cannot show a test of a level, a failed break, or a day that was violent and finished unchanged. Why probability lines mislead is the longer version of that complaint.

A candle is a bucket. Take every trade in an interval. The first is the open. The last is the close. The highest implied probability is the high. The lowest is the low. Do that for every interval and you have a series a technical-analysis toolkit can consume.

Two rules make the rebuild honest:

Go back to the trades, not to the published line. You cannot recover a high from a close series. Any “candle” drawn on top of the venue line is a restyled close. It will look like a chart and contain the same information as the line.

Do not smooth the result. No moving average baked into the series, no outlier clip, no “implausible” wick removed. If a trade printed, it belongs in the bucket. A wick you do not like is a wick that happened. Predm’s reconstruction follows both rules; the OHLC guide is what the four numbers mean once you have them.

A 5-minute candle on a quiet political long-shot is almost always a fiction. A daily candle on a World Cup final in extra time is a real object. The rebuild does not know the difference unless you refuse to draw intervals that do not have enough trades.

Practical starting points:

Daily on any market you would describe to a friend in one sentence. If the daily cannot fill, the market is not chartable yet.

Hourly and below only on books that print through the day. Sports into an event, large political markets into a decision, anything that has been on the front page of the venue for a week.

Minutes for the last stretch before resolution, or not at all. Lower timeframes closer to settlement get noisier, not more precise, because the remaining traders are flattening and squeezing rather than estimating.

Multi-timeframe signals are how you keep those intervals from becoming twelve separate opinions.

Every binary market ends at 0 or 1. That fact is not a reason to ignore the chart. It is a reason to stop treating the last price as the story. The path is what you can trade. The destination is what you settle.

On a rebuilt series you can ask the questions a line cannot: did 50% hold, or did it only print as a wick? Did the high of the day come first (a fade) or last (a grind)? Did the week agree with the hour? Technical analysis for prediction markets is the reading method. This step is just the reminder to look at the path at all.

A 70% close is not a 70% consensus. It is a 70% last price. The last price can be a crowd or it can be four wallets. Holder concentration— Gini, HHI, top-1 and top-5 share, computed per side— is how you tell those two books apart.

This is the step most “how to chart Polymarket” write-ups skip, because it is not a chart. It is the thing that tells you whether the chart is allowed to mean anything. A beautiful bullish engulfing on a YES book owned by one account is a picture of one account.

This is not a ranked list of products. Ranked lists rot the week someone ships a change, and this site does not print claims it cannot check. The questions you can ask of any tool, including Predm, are stable:

Are the candles rebuilt from trades, or restyled from the line? If the vendor cannot say, assume the line.

Is anything smoothed by default? A pretty series is a warning, not a feature.

Can you see more than one timeframe without opening twelve tabs? Prediction markets punish single-timeframe reading.

Can you see concentration next to the chart, per side? A whole-market Gini is the wrong number. The two books can disagree completely.

Is the product a venue, or a read-only terminal? Mixing those two jobs is how a chart quietly becomes a brokerage. Predm is the second kind. You cannot place an order here.

Predm rebuilds OHLC from the raw Polymarket trade feed, draws the candles with nothing smoothed, puts the multi-timeframe board next to them, and computes holder concentration per side. Hyperliquid perps live in the same terminal as ordinary price candles. Two venues are live. Aster, Binance, Kraken and Kalshi are not.

There is no public API. There is no iOS app yet. Account creation is closed while the current beta group is already on the terminal; existing users can log in. Those sentences are also on the homepage, and they are true on this page for the same reason.

The candlestick page is the same argument with the line/candles toggle on a dated snapshot, if you want to see a real market drawn both ways before you touch the terminal.

How do I get candlestick charts on Polymarket?

You do not turn them on in the venue. You rebuild OHLC from the trade feed, or you use a terminal that already does. A browser overlay that draws boxes on the official line is still a line.

Is there a TradingView chart for Polymarket?

TradingView speaks OHLC. The missing piece is a honest OHLC series for the event. Once a series is rebuilt, any OHLC consumer can plot it. The quality of the chart is the quality of the rebuild, not the quality of the skin.

What is the best tool for charting Polymarket in 2026?

The best tool is the one that rebuilds candles from trades, does not smooth them, shows more than one timeframe, and tells you who holds the other side. Predm is built to be that terminal. This page will not rank other products we have not audited this week.

Can I chart Kalshi the same way?

In principle, yes: trades in, OHLC out. Predm does not read Kalshi today. The homepage marks it coming soon, and no chart in this terminal is a Kalshi chart.