Pro
Signals, on the chart and on your phone
The built-in Lock signal, the ones you build yourself, and a notification on your phone the moment any of them happen. Each one is information about what the numbers did, logged with its record.
The built-in signals
◆ SettleMath model: buy signal
When once a market, with 14:00 to 0:30 left, the model's chance of a side sits 3 points or more above that side's asking price after the taker fee (between 5¢ and 95¢). A line on the chart marked Buy Up or Buy Down, and a row on the leaderboard, where it is vetted like every other signal: ahead by more than luck explains over 100+ markets. Drawn on the chart and scored on the Accuracy page; sent to phones once it is vetted.
🔒 Lock
In the last minute (1:00 to 0:08 left), our settlement math - the seconds of the 60-second average already in and where Bitcoin is now, without the market's price - puts one side at 97.5% or more while the market has it at 85¢ to 99.5¢, or at 99% or more above that, up to 99.8¢. Each win pays a few cents and a rare loss costs many wins. (Lock v6, from Oct 28, 3:38 AM CT; v5 needed 98% at 90-99¢ and fired in about one market in five. Earlier versions are on the changes list, with their records as a download.) Once per side per window.
Retired signals (the first Lock and Indicators agree) are listed with their dates on the changes list, where their full records can be downloaded.
33 live readings, each scored on its own record
Fifteen indicators carry points; eighteen more are watched until their record earns them points. Every one is read once a second and every one feeds the SettleMath model, which weighs each reading by what it has actually done against the market's price over the last 14 days, refitted every hour: a reading that starts working is used within the hour, and one that stops is dropped. Nothing is weighted by hand.
Each indicator points Up, Down or neither (inside its dead zone, or without a reading). Every one pointing Up adds its points and every one pointing Down takes them away: the score, from 100 Down to 100 Up, says which way the indicators lean. The points come from each indicator's record on the Accuracy page (10, 5, 2 and 1 minute left): 10 for each one clearly ahead of the market's price so far, 6 for each one a little ahead, none for the rest, shared out of 100 and set again each time the records pass a checkpoint.
| Indicator | Points | Points a way when |
|---|---|---|
| RSI (14), one-minute closes of our price | 0 | Up above 55, Down below 45 |
| Bollinger %B (20, 2) | 0 | Up above 0.6, Down below 0.4 |
| MACD histogram (12, 26, 9) | 0 | beyond 0.001% of the price (about $0.80) |
| 1-minute change | 0 | beyond 0.005% of the price (about $4) |
| 5-minute change | 14 | beyond 0.01% of the price (about $8) |
| Trend: EMA 9 minus EMA 21 of one-minute closes | 0 | beyond 0.002% of the price (about $1.60) |
| Share of the last minute's volume takers bought, 11 exchanges | 0 | Up above 55%, Down below 45% |
| Share of the resting BTC within 0.05% of the price that sits on the bid, on the 5 exchanges whose order books we keep | 14 | Up above 55%, Down below 45% |
| Futures premium over spot, change over a minute | 9 | beyond 0.002% of the price (about $1.60) |
| Coinbase premium: Coinbase's price minus Binance's (dollars against tether), its change over 5 minutes | 0 | beyond 0.005% of the price (about $4) |
| Whale orders of the last 5 minutes, net BTC bought by takers | 9 | 10 BTC or more either way |
| The market's resting orders within 2c of its price on the Yes side | 0 | Up above 55%, Down below 45% |
| The market's Yes price, change over a minute | 9 | 2c or more either way |
| Crowd: the market's trades in the last 2 minutes that took Yes (10 trades or more) | 0 | Up above 55%, Down below 45% |
| The other 15-minute Bitcoin up/down market: its Up price minus this market's Yes price, same window | 0 | 3c or more either way |
| Watched: forced liquidations on the Binance, Bybit and OKX perpetuals in the last 5 minutes, BTC of short positions liquidated (each one pushes the price up) minus long positions liquidated (each pushes it down) | 0 | 5 BTC or more either way |
| Watched: BTC against this window's volume-weighted average price (each second's price weighted by the volume traded on the 11 exchanges), from a minute of volume on | 14 | beyond 0.01% of the price (about $8) |
| Watched: new positions behind the 5-minute move - how much the futures' open interest on Binance, Bybit and OKX grew, counted the way the price moved (shrinking open interest counts as none) | 0 | 100 BTC or more either way |
| Watched: options skew - the nearest expiry at least 6 hours off: its volatility 2% above the price minus 2% below it - its change over 15 minutes (calls getting dearer than puts points Up) | 0 | 0.5 volatility points or more either way |
| Watched: who moved first over the last 2 minutes, Coinbase or the futures, counted the way the price moved (a spot-led move points with it, a futures-led one against it) | 0 | 0.3 or more either way |
| Watched: history vs the market - over the whole trade history (every window since the market began), at this much time left and prices like the one now, how often Yes won minus what those prices said (0 when the market has been priced right there) | 0 | 3 points or more either way |
| Watched: big trades on the market itself in the last 2 minutes - the share of their money taking Yes minus the share taking No (at least 3 big trades' worth) | 0 | 30 points or more either way (65% one way); a big trade costs the taker $1,000 or more |
| Watched: orders pulled from the market's own book - of the contracts resting within 2c of the price 10 seconds ago, how many left the No side minus how many left the Yes side, as a share of all of them (traded or cancelled: the book can't tell which) | 0 | 10 points or more either way, from 500 contracts near the price |
| Watched: the 15-minute ETH market moving first - its Yes price's move over the last minute minus this market's own, same window | 0 | 3c or more either way |
| Watched: the crowd by contracts - of the contracts the market's takers took in the last 2 minutes, the share on Yes. Every contract is the same bet from either side, so the cheap side's many small orders don't tilt it, and neither does the price | 0 | above 55% or below 45%, from 10 trades |
| Watched: big money against small trades - of the contracts in the last 2 minutes' trades costing $100 or more, the share on Yes, minus the same share for trades under $5. Where the big money goes against the many small orders, the big money's side has won more often than its price so far | 0 | 20 points or more either way |
| Watched: the market's own price velocity - its Yes price now minus 10 seconds ago | 14 | 1.5c or more either way |
| Watched: where Bitcoin sits in its range of the last 5 minutes, from the low (0%) to the high (100%) - a breakout reading | 9 | above 80% or below 20%, from a range of about $8 or more |
| Watched: the move accelerating - the last minute's change minus a fifth of the last 5 minutes' | 9 | about $2.40 or more either way |
| Watched: the fast exchanges ahead of our price - the 5-second move of Coinbase's book and the Binance and OKX futures, minus our own price's. The market moves with those fast prices | 0 | $3 or more either way |
| Watched: the other coins' 15-minute markets (SOL, XRP, DOGE and more) moving first - their average Yes move over the last minute minus this market's own | 0 | 3c or more either way, from 2 markets |
| Watched: the final-minute arithmetic against the price - our chance of Yes from the seconds already averaged and our price alone, without the market's price, minus the market's price | 0 | 5 points or more either way |
| Watched: the windows just before this one - how many in a row settled Up, or Down - pointing the way of the streak, a test of whether streaks carry on | 0 | any streak: 1 window or more |
These levels are starting values while the record builds. Every signal is logged, and the Accuracy page shows how often the side each one pointed to won, next to what the market was already pricing when it fired. They describe what the numbers did; they are not advice.
The top-3 signal bot: a bot that trades whichever of the leaderboard's top 3 fires first in each market, every trade since it started (a live-price simulation). See its record →
Your own signals
On the Live page, the Your signals panel lets you combine up to six conditions on the page's numbers: the market's price, the model, the flip chance, RSI, volume, order flow, time left and more. A line is drawn on the chart every time they all line up, and with an account each signal keeps its record for the last 90 days: in how many windows it fired, and how often the side it pointed to won.
Browser notifications
Get each signal as a notification on your phone or computer the moment it fires, even with the page closed: the Lock signal, and your own signals that have "Also notify me" ticked. One switch per browser.
On an iPhone or iPad, notifications work once SettleMath is on your Home Screen: in Safari tap Share, then Add to Home Screen, open SettleMath from the new icon, and turn this on there. On Android and computers, any current Chrome or Firefox (or another up-to-date browser) works as it is.
Discord
Have the signals posted in a channel of your own Discord server: the Lock signal if you choose it, and your own signals that have "Also notify me" ticked.
Sign in with a Pro plan to send signals to your Discord.
Telegram
Sign in with a Pro plan to connect Telegram.
The bot only reads /start, /stop and /help. Send /stop to it, or press Disconnect here, to stop every message.