Half a Clock, Twice the Trouble

The short version: My opponent was rated 808 points higher and voluntarily threw away half the clock. This seemed sporting. I responded by taking one bishop on a four-move tour, assembling an enormous time advantage, and discovering that neither of those things counts as material.

This is game two of the Monthly Classical Arena and the immediate sequel to Not Going Down Without a Fight.

The tournament campaign, at the moment I opened this review:

Zero wins. Zero draws. Two losses. No points, folks.

But—and this is important—we were not last. There were 244 players below ThreeSadPandas in that snapshot. By the time Stockfish finished, the live table had admitted seven more players and shifted me to 1,197th of 1,440. The numbers moved. The 0 remained magnificently stable.

The arena eventually finished with 1,794 players. ThreeSadPandas landed in 1,512th place: zero wins, zero draws, two losses, zero points, and 282 players behind me. The snapshots were temporary. Their strategic meaning survived certification: not last.

This review uses the same reproducible method as the first game. Stockfish 18 examines three candidates before every played move, then searches the resulting position separately. The complete report and exact clocked PGN are preserved in the repository.

As usual, the full evaluation ledger waits until after the story. You deserve the opportunity to develop an unreasonable amount of confidence in the clock.

Replay the clock advantage

The official Lichess board is oriented from White’s side.

The original remains available as UnumTsci on Lichess.

The complete game score

Watch the clocks carefully. Black begins with ten minutes because Cheron berserked. In an arena, berserk halves the starting time, removes the increment, and offers an extra tournament point for a win after enough moves. White keeps the full 20+10.

[Event "Monthly Classical Arena"]
[Site "https://lichess.org/UnumTsci"]
[Date "2026.09.04"]
[Round "-"]
[White "ThreeSadPandas"]
[Black "Cheron"]
[Result "0-1"]
[GameId "UnumTsci"]
[UTCDate "2026.09.04"]
[UTCTime "19:35:07"]
[WhiteElo "1166"]
[BlackElo "1974"]
[WhiteRatingDiff "+0"]
[BlackRatingDiff "+1"]
[BlackBerserk "true"]
[Variant "Standard"]
[TimeControl "1200+10"]
[ECO "B20"]
[Opening "Sicilian Defense"]
[Termination "Normal"]

1. e4 { [%clk 0:20:00] } 1... c5 { [%clk 0:10:00] }
2. Bb5 { [%clk 0:19:33] } 2... a6 { [%clk 0:09:56] }
3. Ba4 { [%clk 0:19:39] } 3... b5 { [%clk 0:09:52] }
4. Bb3 { [%clk 0:19:43] } 4... c4 { [%clk 0:09:50] }
5. c3 { [%clk 0:19:40] } 5... cxb3 { [%clk 0:09:47] }
6. Qxb3 { [%clk 0:19:48] } 6... Bb7 { [%clk 0:09:04] }
7. Nf3 { [%clk 0:19:51] } 7... Bxe4 { [%clk 0:08:59] }
8. Ng5 { [%clk 0:19:53] } 8... Bg6 { [%clk 0:08:53] }
9. O-O { [%clk 0:19:56] } 9... e6 { [%clk 0:08:39] }
10. Re1 { [%clk 0:19:59] } 10... Qxg5 { [%clk 0:08:35] }
11. d4 { [%clk 0:19:50] } 11... Qd8 { [%clk 0:07:59] }
12. Bf4 { [%clk 0:19:36] } 12... Be7 { [%clk 0:07:39] }
13. Nd2 { [%clk 0:19:42] } 13... Nf6 { [%clk 0:07:30] }
14. Rad1 { [%clk 0:19:36] } 14... d5 { [%clk 0:07:25] }
15. a4 { [%clk 0:19:40] } 15... bxa4 { [%clk 0:07:02] }
16. Qxa4+ { [%clk 0:19:07] } 16... Qd7 { [%clk 0:07:00] }
17. Ra1 { [%clk 0:18:45] } 17... O-O { [%clk 0:06:52] }
18. Qd1 { [%clk 0:18:28] } 18... Nc6 { [%clk 0:06:43] }
19. Ra4 { [%clk 0:18:14] } 19... h6 { [%clk 0:06:40] }
20. b4 { [%clk 0:18:01] } 20... Rfc8 { [%clk 0:06:16] }
21. Qc1 { [%clk 0:17:42] } 21... Nxd4 { [%clk 0:06:03] }
22. Be5 { [%clk 0:17:31] } 22... Nc2 { [%clk 0:05:33] }
23. Rd1 { [%clk 0:16:55] } 23... Qxa4 { [%clk 0:05:28] }
24. Nb1 { [%clk 0:16:19] } 24... Nxb4 { [%clk 0:05:18] }
25. Na3 { [%clk 0:15:53] } 25... Na2 { [%clk 0:05:04] }
26. Qd2 { [%clk 0:15:14] } 26... Qxa3 { [%clk 0:04:52] }
27. Ra1 { [%clk 0:14:51] } 27... Ne4 { [%clk 0:04:38] }
28. Qd3 { [%clk 0:14:29] } 28... Nexc3 { [%clk 0:04:29] }
29. Qd4 { [%clk 0:14:10] } 29... Ne2+ { [%clk 0:04:26] }
30. Kf1 { [%clk 0:14:14] } 30... Nxd4 { [%clk 0:04:24] }
31. Bxd4 { [%clk 0:14:22] } 31... Bd3+ { [%clk 0:04:18] }
32. Ke1 { [%clk 0:14:30] } 32... Bb4+
{ [%clk 0:04:07] } 33. Kd1 { [%clk 0:14:36] } 33... Qb3#
{ [%clk 0:04:05] } 0-1

Before move one: a very generous stranger

Cheron entered at 1974. I entered at 1166. That is an 808-point gap.

Then Black pressed the berserk button.

Instead of 20+10, Cheron accepted 10+0. I retained twenty minutes and received ten more seconds after every move. On paper, this is the practical equalizer: the stronger player must solve every problem in half the time while the lower-rated player can make the position unpleasant and wait for the clock to help.

The missing clause is that White must first create problems.

The bishop tour

1.e4 c5 2.Bb5 a6

The game begins with the Sicilian Defense. 2.Bb5 is playable, although Stockfish prefers ordinary development with 2.Nf3. Black immediately asks the bishop to explain itself.

After 2…a6, the bishop has a safe and simple retreat: 3.Be2. Stockfish evaluates that position at approximately −0.05, which is another way of saying nothing has happened.

I choose 3.Ba4??.

This is the game’s decisive mistake, even though no capture occurs yet. Stockfish falls from essentially equal to −3.25 because Black has a forcing net:

3.Ba4 b5 4.Bb3 c4

The bishop is on b3. Black’s c-pawn attacks it from c4. The exit squares are gone:

• a4 is controlled by the pawn on b5;
• c2, d1, and a2 are occupied by White’s own pieces; and
• c4 contains the pawn doing the attacking.

This is not a one-move blunder. It is a three-move corridor that ends in a wall.

4.Bb3 c4 5.c3 cxb3 6.Qxb3

I play 5.c3, perhaps hoping that building a small white pawn neighborhood around the bishop will make Black feel awkward about taking it.

Black feels fine.

Stockfish prefers 5.d4, not because it saves the bishop—it cannot—but because it claims the center and gets compensation while Black spends a move collecting. After 5…cxb3, the bishop tour is complete:

Bb5–a4–b3–off the board.

White has used four of the first five moves on one bishop and lost it. Black has used a-, b-, and c-pawns to do the trapping, so there are loose squares and a lead in development to work with. The position is bad, but at least it has a practical instruction: open the center, develop quickly, and make the extra piece work for its salary.

Also, clocks:

• White: 19:48;
• Black: 9:47.

We are down a bishop and up ten minutes.

If one minute equals roughly a third of a pawn, everything is fine. Please do not inspect that conversion rate.

Finding a second donation

6…Bb7 7.Nf3? Bxe4

Black develops the bishop onto the long diagonal and immediately attacks e4. I develop with 7.Nf3, but the e-pawn is loose. Black takes it.

The engine prefers 7.d3, protecting the center while opening the dark-squared bishop. After 7…Bxe4, Black has an extra bishop and an extra pawn.

I answer with activity.

8.Ng5 Bg6 9.O-O e6

8.Ng5 attacks the bishop and points toward f7. Black retreats. I castle. The king is safe, the rook reaches f1, the queen is active on b3, and Black has not castled.

This is the visual case for hope. White’s pieces are beginning to point at the kingside. Black has spent pawn moves trapping the bishop. The berserk clock has fallen to 8:39. If I can open the center, Cheron will have to calculate with no increment.

The knight on g5, however, has to come home.

10.Re1?? Qxg5

After 9…e6, Stockfish wants 10.d4 or 10.Nf3. Either move begins repairing the center; the second simply removes the knight from danger.

I play 10.Re1, placing the rook opposite the king and adding another piece to the idea of central pressure.

Black’s queen travels from d8 to g5 and takes the knight.

The diagonal is clear. There is no combination after the capture. There is only one more white piece missing from the board.

Stockfish gives 10.Re1?? a loss of 0.87 pawns relative to the best defense and evaluates the result at −5.63. That number is important because the position was already badly lost. The move did not create the emergency; it declined the best chance to contain it.

Black now has both of White’s kingside minor pieces. White has a very impressive clock.

The twelve-minute comeback

11.d4 Qd8 12.Bf4 Be7 13.Nd2 Nf6 14.Rad1 d5

Now the moves I needed several turns ago begin to appear. d4 claims the center. The remaining bishop develops to f4. The queenside knight reaches d2. A rook comes to the open file.

Black retreats the queen, develops, and builds a stable center.

Objectively, this is not a comeback. Stockfish stays between roughly −5.4 and −7.0. Visually, though, White has finished development, Black has only seven and a half minutes, and the pieces are finally facing the correct direction.

We will call that a comeback because this is my website and the evaluation table has not appeared yet.

15.a4! bxa4 16.Qxa4+

Here White plays two excellent moves.

15.a4 is Stockfish’s first choice. It attacks the queenside chain and opens lines before Black can finish coordinating. After 15…bxa4, 16.Qxa4+ is again the engine’s first choice.

Check forces a response. The queen becomes active. Black’s exposed king has to spend a move on 16…Qd7.

The engine still says approximately −6.55. We are not asking the engine right now.

16…Qd7 17.Ra1 O-O 18.Qd1 Nc6 19.Ra4 h6 20.b4

The rook swings to a1, then lifts to a4. The b-pawn advances beside it. White has a queen, two rooks, a bishop, a knight, space on the queenside, and twelve more minutes than Black.

Black has the pieces that disappeared earlier, which feels like accounting.

After 20.b4, Cheron has 6:16. I have 18:01. The clock gap is now 11 minutes and 45 seconds.

This is where the berserk strategy can become real. The stronger player needs positions with easy moves. The lower-rated player needs to keep pieces active, avoid exchanges, and present several plausible answers. A huge clock edge does not attack by itself; it amplifies the difficulty you manage to create.

My rook on a4 looks active. It is also standing on the same diagonal as Black’s queen on d7.

That will matter shortly.

The board starts collecting debts

20…Rfc8 21.Qc1? Nxd4

Black places the rook on the c-file. I move the queen to c1, lining up behind the bishop and imagining pressure.

Stockfish prefers 21.Qa1, 21.Ra1, or 21.Nb3. All three attend to Black’s active pieces. 21.Qc1? allows 21…Nxd4, and the knight lands in the center with threats in several directions.

The evaluation reaches −7.22.

22.Be5 Nc2 23.Rd1? Qxa4

I develop the bishop to e5. Black’s knight jumps to c2. It attacks the rook on a1 and enters White’s position with tempo.

The rook on a4 is still sitting on the queen’s diagonal.

Stockfish wants 23.Ra5, moving that rook away and limiting the damage. I play 23.Rd1?, moving the other rook.

Cheron plays 23…Qxa4.

The queen crosses d7–c6–b5–a4 and takes the rook. White is now down two minor pieces, a rook, and pawns.

But Black has only 5:28.

You see the issue with the clock thesis.

24.Nb1 Nxb4 25.Na3 Na2 26.Qd2 Qxa3

My remaining knight tries to reroute through b1 and a3. Black’s other knight takes b4. Then the centralized knight reaches a2, and the black queen takes the knight on a3.

The board is no longer cluttered with White’s minor pieces. This does improve visibility.

Stockfish passes −10. Black has under five minutes, but the position is simplifying in exactly the wrong direction. Cheron does not need to solve a mystery. Cheron needs to keep taking things that are attacked.

One last chance to choose the shape of the loss

27.Ra1 Ne4 28.Qd3?

The last rook attacks the knight on a2. Black centralizes the other knight with 27…Ne4.

Here Stockfish prefers 28.Qxa2!:

28.Qxa2 Qxa2 29.Rxa2

White can trade queens and eliminate the advanced knight. The resulting ending is still completely lost—about −9.9—but the immediate attack disappears and Black must convert material.

I play 28.Qd3? instead, preserving queens and keeping the position complicated.

This is philosophically consistent with exploiting a berserked opponent. It is also tactically dangerous when the opponent is the one with all the pieces.

28…Nexc3 29.Qd4? Ne2+

One knight captures on c3. I centralize the queen with 29.Qd4? and point it toward the kingside.

Stockfish wants 29.Qf1, keeping the queen away from the knight’s tempo. Black finds 29…Ne2+.

It is a fork with check. The knight attacks the king and the queen on d4. White must answer the check; then Black takes the queen.

At this point, the practical comeback has one remaining ingredient: perhaps a 1974-rated player with 4:26 will forget what a fork is.

Cheron remembers.

30.Kf1? Nxd4

The best king move is 30.Kh1, which loses the queen but avoids an immediate forced mate. Stockfish evaluates it around −13.56. This is not salvation. It is a longer administrative process.

I choose 30.Kf1?.

Black takes the queen. Stockfish announces mate in six.

The clock says:

• White: 14:14;
• Black: 4:24.

Ten minutes and an entire future remain on my clock. Six moves remain in the position.

Down with honor

31.Bxd4 Bd3+ 32.Ke1 Bb4+ 33.Kd1 Qb3#

I take the knight that captured my queen. Cheron answers with a bishop check. The king retreats. The other bishop checks. The king retreats again.

Then 33…Qb3#.

The queen, both bishops, rook, and knight coordinate around the king. There is no block, capture, or flight square. Cheron finishes with 4:05.

I finish with 14:36.

This is the opposite of the first tournament game. There is no pawn race, no last-rank promotion, and no newborn queen asking for a ceremonial moment. This time the honorable act is smaller: do not resign in the middle of the combination. Make the stronger player place every final piece correctly. Let the checkmate exist on the board.

The panda goes down with self-respect and an absolutely irresponsible quantity of unused clock.

Now we may look at the engine ledger

Every score is from White’s perspective. Negative favors Black. Mate scores replace decimals once Stockfish sees a forced finish.

There is no hidden winning phase here. Unlike the first game, the hopeful section was a narrative misdemeanor. White’s best stretch after the opening consisted of finding 15.a4 and 16.Qxa4+, gaining activity without repairing the material deficit.

The main evaluation swing occurred on move three. Everything after that was conversion, resistance, or further donation.

What I did well

I castled

After losing the bishop and e-pawn, 9.O-O was Stockfish’s first choice. The position needed king safety and rook activation more than it needed an exotic counterattack.

I found the queenside break

15.a4! and 16.Qxa4+ were both best moves. They opened a line, activated the queen, and forced Black to answer a check. In a bad position, forcing the opponent to respond is how practical chances begin.

I made Black finish

Cheron was rated 1974 and played the tactical finish correctly. I did not convert a lost position into a victory, but I also did not replace the final calculation with a resignation. The mating pattern is now available for study because it happened on the board.

My rating survived

The PGN records a White rating change of +0.

Score: zero. Rating lost: zero. Last place: no.

We take our trophies where the regulations permit.

What I need to fix

Trace the bishop’s entire retreat path

After 2…a6, choosing between Ba4 and Be2 is not only about the next square. Before retreating, ask what happens after the opponent gains another tempo with …b5, then whether a pawn push such as …c4 can close the final exit.

Training question: if this piece is attacked twice in succession, where does it end up?

Run checks, captures, and threats for the opponent

10…Qxg5, 23…Qxa4, and 29…Ne2+ are all visible when the scan starts from Black’s forcing moves rather than White’s plans.

Training question: after I make this move, what checks and captures become legal for my opponent?

Spend classical time at irreversible moments

White finished with more than fourteen minutes. The critical choices—3.Ba4, 10.Re1, 23.Rd1, 28.Qd3, and 29.Qd4—received nowhere near the time their consequences deserved.

A ten-minute advantage over a berserked opponent is not a reason to play faster. It is permission to make the opponent’s gamble expensive.

Use clock pressure by preserving decisions

To exploit a short clock, keep threats alive, avoid unnecessary simplification, and offer the opponent multiple credible replies. But material safety comes first. Complication without enough pieces is not pressure; it is additional surface area for tactics.

When lost, choose the most difficult loss

After 27…Ne4, 28.Qxa2! forces a queen trade and removes a dangerous knight. The ending remains lost, but it makes Black demonstrate technique instead of attack. Practical resistance is not always keeping queens. Sometimes it is removing the opponent’s easiest route.

Reproducible engine record

The run used:

• Stockfish 18;
• python-chess 1.11.2;
• 4 threads;
• 512 MB hash;
• 5,000,000 nodes for each search;
• MultiPV 3 before every played move;
• a separate 5,000,000-node search after every played move; and
• principal variations limited to 12 plies in the saved report.

Scores are normalized to White’s perspective. Multi-threaded searches may vary by a few centipawns between runs; none of the conclusions here depends on noise at that scale.

The repository-owned analysis harness produced the report:

#!/usr/bin/env python3
"""Produce a reproducible, move-by-move Stockfish report for a PGN game."""

import argparse
import json
import shutil
from pathlib import Path

import chess
import chess.engine
import chess.pgn


def score_for_white(info):
    return info["score"].pov(chess.WHITE).score(mate_score=100000)


def display_score(info):
    score = info["score"].pov(chess.WHITE)
    if score.is_mate():
        mate = score.mate()
        return f"M{mate}" if mate is not None and mate >= 0 else f"-M{abs(mate or 0)}"
    value = score.score()
    return f"{(value or 0) / 100:+.2f}"


def pv_to_san(board, pv, length=12):
    variation = board.copy()
    moves = []
    for move in pv[:length]:
        if move not in variation.legal_moves:
            break
        moves.append(variation.san(move))
        variation.push(move)
    return " ".join(moves)


def main():
    parser = argparse.ArgumentParser()
    parser.add_argument("pgn", type=Path)
    parser.add_argument("output", type=Path)
    parser.add_argument("--engine", default=shutil.which("stockfish") or "stockfish")
    parser.add_argument("--nodes", type=int, default=500000)
    parser.add_argument("--multipv", type=int, default=3)
    parser.add_argument("--threads", type=int, default=4)
    parser.add_argument("--hash", type=int, default=512)
    args = parser.parse_args()

    with args.pgn.open() as source:
        game = chess.pgn.read_game(source)
    if game is None:
        raise SystemExit("No PGN game found")

    engine = chess.engine.SimpleEngine.popen_uci(args.engine)
    engine.configure({"Threads": args.threads, "Hash": args.hash})
    report = {
        "game": dict(game.headers),
        "method": {
            "engine": engine.id,
            "python_chess": chess.__version__,
            "nodes_per_search": args.nodes,
            "multipv": args.multipv,
            "threads": args.threads,
            "hash_mb": args.hash,
            "score_perspective": "White; positive favors White",
            "mate_score_centipawns": 100000,
            "pv_length_plies": 12,
        },
        "positions": [],
    }

    try:
        board = game.board()
        for ply, move in enumerate(game.mainline_moves(), start=1):
            mover = board.turn
            move_number = board.fullmove_number
            played_san = board.san(move)
            candidates = engine.analyse(
                board,
                chess.engine.Limit(nodes=args.nodes),
                multipv=args.multipv,
            )
            candidate_rows = []
            for rank, info in enumerate(candidates, start=1):
                candidate_rows.append(
                    {
                        "rank": rank,
                        "move": board.san(info["pv"][0]),
                        "score": display_score(info),
                        "score_cp_white": score_for_white(info),
                        "depth": info.get("depth"),
                        "seldepth": info.get("seldepth"),
                        "nodes": info.get("nodes"),
                        "pv": pv_to_san(board, info["pv"]),
                    }
                )

            best_cp = candidate_rows[0]["score_cp_white"]
            best_move = candidates[0]["pv"][0]
            board.push(move)
            after = engine.analyse(board, chess.engine.Limit(nodes=args.nodes))
            after_cp = score_for_white(after)
            loss = best_cp - after_cp if mover == chess.WHITE else after_cp - best_cp

            report["positions"].append(
                {
                    "ply": ply,
                    "move_number": move_number,
                    "side": "White" if mover == chess.WHITE else "Black",
                    "played": played_san,
                    "best_move": candidate_rows[0]["move"],
                    "is_best": move == best_move,
                    "centipawn_loss": max(0, loss),
                    "evaluation_after": display_score(after),
                    "evaluation_after_cp_white": after_cp,
                    "after_depth": after.get("depth"),
                    "after_seldepth": after.get("seldepth"),
                    "after_nodes": after.get("nodes"),
                    "best_reply_line_after": pv_to_san(board, after.get("pv", [])),
                    "candidates_before": candidate_rows,
                }
            )
    finally:
        engine.quit()

    args.output.parent.mkdir(parents=True, exist_ok=True)
    args.output.write_text(json.dumps(report, indent=2) + "\n")


if __name__ == "__main__":
    main()

The preserved source files are:

• analysis/stockfish/UnumTsci.pgn — the exact Lichess PGN with clocks;
• analysis/stockfish/UnumTsci-stockfish18.json — the complete five-million-node report; and
• scripts/analyze_game.py — the reusable analysis harness.

The PGN and report remain internal source evidence rather than public downloads. The script is displayed through the repository-restricted shortcode. Stockfish is free, open-source software; this site invokes a local installation and does not redistribute the engine binary. The evergreen explanation is in What Stockfish Is and How I Use It.

The honest verdict

My opponent gave me half the clock. I could not turn that gift into hard decisions because I gave away the bishop on move five and the knight on move ten. From there, Cheron had an easier position, more material, and enough time.

The lesson is not “ratings decide games.” A rating did not play 3.Ba4. A rating did not leave the knight on g5 or the rook on a4. Those were concrete mistakes with concrete repairs: map retreat squares, scan forcing moves, and spend time before irreversible decisions.

After two games, the Monthly Classical Arena score reads:

0 wins. 0 draws. 2 losses. 0 points.

Final place: 1,512th of 1,794.

Not last.

Folks, there is still room below us.