You can play rock paper scissors online in your browser against a transparent opponent and finish a complete run with exactly 1,000 points by choosing Scissors, then Paper, then Rock against the opponent's disclosed counters of Paper, Rock, and Scissors. The match is deterministic, runs entirely in the browser, and does not depend on a remote service, hidden randomizer, language model, or downloaded app. Every round shows the prediction source, the predicted player move, and the counter the opponent has already queued, which makes each choice an inspectable decision rather than a guess. Wins are worth 300 points, losses subtract 100 without dropping the score below zero, and reaching three wins adds a 100-point completion bonus, so three wins in a row total 900 plus 100 for exactly 1,000. The whole run is reproducible because the opponent reads only the moves recorded in the current browser session and uses a counting rule with a fixed tie priority. Two losses close the run, ties do not change wins, losses, or score but still enter history, and Restart clears only the local state and returns to the same fixed opening prediction.
About Rock Paper Scissors Online
The Rock Paper Scissors Online game is a browser-only match against a transparent, deterministic prediction rule. The opponent is not a person, not a language model, and not a hidden random number generator. It reads the moves you have played in the current run, decides what it expects you to play next using a fixed counting rule, and queues the move that beats that prediction before you make your choice. Because the queued counter and the prediction source are visible, you can decide whether to follow the suggestion, try to break the pattern, or play the verified winning route.
This setup turns the page into a small prediction puzzle rather than a sealed battle. It also makes the game fully reproducible: the same sequence of moves produces the same opponent response every time, on any browser, with no account, no upload, and no third-party request. You can play rock paper scissors directly in your browser tab and inspect the rule as you go.
How the Opponent Predicts Your Next Move
The opponent follows a single, fixed rule that has three parts. On the opening round, it uses a baseline prediction of Rock and queues the move that beats Rock, which is Paper. After at least one move has been played, the opponent first checks whether your latest move appeared earlier in the run followed by another move. If such a transition sample exists, it predicts the most frequent move that historically followed your latest move. If no usable transition sample exists, it predicts the most frequent move in your entire move history.
Equal counts always resolve in the disclosed order Rock, then Paper, then Scissors. Once the prediction is fixed, the opponent queues the move that beats it before you choose. Because a legal repeated choice is treated as a new learning sample rather than an ignored duplicate, even playing the same move twice changes the opponent's view of your pattern. Ties also enter history because a tie can shift which transition or frequency counts dominate a later round, but ties change neither your wins, your losses, nor your score.
Scoring Rules and Run Conditions
Every round resolves into one of three outcomes: a win, a loss, or a tie. Wins and losses move the score; ties are recorded but stay neutral. The exact rules, taken directly from the verified product contract, are summarized in the table below.
| Outcome | Score change | When it applies |
|---|---|---|
| Win | +300 | Your move beats the opponent's queued move |
| Loss | −100 (floor 0) | Your move loses to the opponent's queued move |
| Tie | 0 | Both moves are identical |
| Completion bonus | +100 | Awarded the moment your third win is recorded |
| Run termination | Final state frozen | Two losses close the run; later choices are disabled |
A run ends in one of two ways. Either you reach three wins, which freezes the choice buttons and locks in the final score of exactly 1,000 points, or you reach two losses, which also freezes the interface and prevents further legal input. There is no overtime and no extra round after a third win or a second loss.
How to Play Rock Paper Scissors Online
The interface shows the current prediction source, the predicted player move, the queued opponent move, the running totals of wins, losses, and ties, a list of recent rounds, and a short plain-language explanation of the rule. To start a run, follow these steps.
- Open Rock Paper Scissors Online in your browser and read the visible prediction source, predicted player move, and queued opponent counter before you make any choice. On the opening round these read baseline, Rock, and Paper.
- Pick a move by clicking one of the three large buttons or by pressing 1 or R for Rock, 2 or P for Paper, and 3 or S for Scissors. The opponent records the choice immediately and updates the score, totals, and recent rounds.
- Watch how the prediction source changes after your first move. If the opponent now lists a transition sample, it means your latest move appeared earlier in the run; if it lists history frequency, no transition sample exists and the opponent fell back to your overall most frequent move.
- Decide whether to follow the verified winning route, experiment with a different sequence, or test how a specific pattern changes the opponent's next counter. Each legal choice is a new learning sample, so even repeated moves affect future predictions.
- If you record a loss or a tie, the run continues. If you record a second loss, the run closes and the interface freezes; press Restart to clear the local state and return to the fixed opening prediction.
- If you record three wins in a row, the run ends with a completion bonus, the score settles at exactly 1,000, and the choice buttons stop accepting input. Restart again to explore a different history.
The Verified 1,000-Point Winning Route
The verified perfect route is three moves in order: Scissors, then Paper, then Rock. Against the opponent's disclosed counters, this sequence produces three consecutive wins.
On the opening round the baseline predicts Rock and queues Paper, so Scissors beats Paper for a first win worth 300 points. After that move, the opponent reads your latest move as Scissors; no earlier Scissors transition exists yet, so it falls back to your overall move frequency. With only Scissors in history, the opponent predicts Scissors and plays Rock, which means Paper beats Rock for a second 300-point win. With two moves now in history, the opponent again checks transitions from your latest move, Paper; none exist, so it predicts the most frequent move overall. Scissors and Paper are tied at one each, and the disclosed Rock-first tie priority resolves the tie toward Paper, so the opponent queues Scissors (the move that beats Paper). Your third move is Rock, which beats Scissors for a third 300-point win. The completion bonus then adds 100 points.
The arithmetic is straightforward: 300 + 300 + 300 + 100 = 1,000. Every step of this route can be replayed by hand, because the prediction rule is public and the counters are visible. The disclosed failure route, by contrast, is Rock, Paper, Paper, Paper, which resolves as loss, tie, tie, loss and demonstrates both the baseline and a learned transition counter before the second loss closes the run.
Controls, Shortcuts, and Restart
The interface accepts input in two ways: mouse or touch on the three large Rock, Paper, and Scissors buttons, and the keyboard shortcuts 1 or R for Rock, 2 or P for Paper, and 3 or S for Scissors. Both routes are equally valid; the opponent does not distinguish between them. An invalid key, such as pressing Q on the choice prompt, is an atomic no-op and does not consume a turn, advance history, or alter the score.
The Restart control clears only the current game's local state. It resets the prediction source to baseline, the predicted move to Rock, the queued counter to Paper, and the wins, losses, ties, and score to zero. It does not change any persistent setting, because there are none. After Restart, the next choice behaves exactly like the very first round of a fresh run.
Privacy and What the Opponent Does Not Do
The opponent calls no model, no remote API, and no random generator. It does not write to a server, log moves outside the current browser session, infer anything about the player beyond the displayed move list, or personalize behavior beyond the visible counters. The word learning here has a narrow, technical meaning: deterministic counting within the current run, the same way a calculator counts the keys you press.
Because the rule is fully disclosed and reproducible, familiarizing yourself with the fixed route makes replay straightforward and lets you focus on small variations rather than guessing. If you want to study how a specific history changes the opponent's prediction source and queued counter, use Restart to clear the run and replay a different move sequence. If you want the verified 1,000-point finish, follow the exact Scissors, Paper, Rock route and stop when the interface freezes. The score is not framed as a psychological, cognitive, educational, or employment assessment; it is simply a deterministic total of the published scoring rules.
For a deeper look, see Rotate the Picture Puzzle: 18 Turns to 1,000 Points.