Teacher copy · Velocity Arena
There is no single set of right numbers. Every team's equation is different, so every team's times are different — and different is correct. This key gives you one fully worked bot on the left and, for each step, the rule that decides whether any team's work is right on the right. Grade against the rule, not the numbers.
let driveMs. Different targets → different times → same speed. Watch for teams trying to change their speed instead; the speed does not change.distance = 38 × time − 7.
Speed = 38 cm/s · start-up = −7 cm.
Speed = the number in front of time (the slope). Start-up = the second number, and it should be negative.
"Speed" here is the cm-per-second from the equation — not their Speed stat. If a team wrote their Speed stat (a small number like 5), that's the wrong number.
time = (target − start-up) ÷ speed
= (target − (−7)) ÷ 38 = (target + 7) ÷ 38 seconds.
Then to program it: seconds × 1000 = ms.
Subtract the start-up first, then divide by speed. The sign is the trap — start-up is negative, so subtracting it adds (target + 7). target − 7 dropped the double negative.
Then the ms step: multiply the seconds by 1000. A time of 3.6 s is driveMs = 3600, not 360 — the other classic slip.
time = (130 + 7) ÷ 38 = 137 ÷ 38 = 3.605 s → × 1000 = 3605 ms.
Set driveMs = 3605, download, press A. Model lands ~130. Miss is a few cm at most.
Their time = (their target + their start-up-flipped) ÷ their speed, × 1000 for the ms. Plug their numbers; it should match, and driveMs should be that many ms.
Miss = target − where it actually stopped, a plain +/− number. 2–4 cm is a good result.
driveMs is milliseconds, so 3.6 s is 3600, not 360; (2) editing the wrong line — only let driveMs changes, marked ★ in the code. Everything else stays put, including SPEED, which they set once to match their Distance Lab.If the bot overshoots every target, the real speed is a little higher than 38 (or the start-up is off). Nudge one number and re-aim.
The "one thing" must be named: a better speed, a better start-up, or a straighter drive. "We changed it" isn't an answer.
Across the rounds the miss should shrink. If it doesn't, ask what number they changed and why.
time = (107 + 7) ÷ 38 = 114 ÷ 38 = 3.0 s → 3000 ms. Model lands ~107 — a strong hit.
Everyone aims at the same surprise target and everyone programs the exact time, so it is fair — the winner has the sharpest model + measuring, not the luckiest grid.
Best of 2, closest miss wins. The spread between teams is mostly a few cm of mechanical noise, so best-of-2 and one neutral ref measuring at the line keeps it clean. Crown a champion.