Guessing how much a giraffe weighs or how many pieces are in a LEGO castle feels like pure luck the first time. It isn't. Estimation is a skill, and a handful of simple habits will move your guesses from “wildly off” to “surprisingly close” within a few rounds. This article covers the techniques that work best, then looks at how Guessit.io scores your answers, because knowing the scoring rule changes which number you should type.
An anchor is a reference value you know well and can compare against. The best estimators don't know more facts than everyone else. They have a small set of reliable anchors and they reason from them. A few useful ones:
When a new item appears, ask: “Is it bigger or smaller than my nearest anchor, and by how much?” That turns a blind guess into a comparison, and people are much better at comparing than at producing numbers from nothing.
The single biggest source of lost points is being off by a factor of ten or more. Before you think about the exact number, decide how many digits it should have. Is this cat closer to 400 g, 4,000 g or 40,000 g? Once the number of zeros is right, refining the first digit is easy.
This matters most in Guess The Weight, where answers range from a thousandth of a gram for a flea to trillions of grams for a pyramid. Saying the number out loud in a bigger unit (“four kilos”, “ten tonnes”) and converting at the end helps avoid typing one zero too many or too few.
Physicist Enrico Fermi was famous for estimating quantities that seemed impossible to know by splitting them into smaller questions he could answer. The same trick works here:
Each sub-estimate will be a little wrong, but the errors tend to partially cancel out, so the final result is usually better than a single gut feeling.
Every Guessit.io round awards up to 1000 points using the same formula:
points = 1000 × e−3 × error, where error = |guess ÷ answer − 1|.
In plain words, only the percentage you are off matters, and points drop off quickly but smoothly.
| Your guess compared to the answer | Points |
|---|---|
| One tenth (×0.1) | 67 |
| Half (×0.5) | 223 |
| 20% too low (×0.8) | 549 |
| 10% too low (×0.9) | 741 |
| Exact | 1000 |
| 10% too high (×1.1) | 741 |
| 20% too high (×1.2) | 549 |
| One and a half times (×1.5) | 223 |
| Double (×2) | 50 |
| Triple (×3) | 2 |
Look at the two ends of the table. Guessing half the answer still earns 223 points, but guessing double earns only 50. Guessing ten times too low (67 points) actually beats guessing twice too high. A guess that is too low can never be more than 100% off, while a guess that is too high can be off by 200%, 500% or more.
That asymmetry has a practical consequence. Suppose your gut says “about 1,000”, but honestly the answer could be anywhere from 500 to 2,000. If you type 1,000, your average score over many such rounds is about 420 points. If you shade your guess down to about 800, the average rises to roughly 440. The gain is small for a single round, but it adds up over a long game.
The effect is much stronger when you are choosing between two very different numbers. If you think the answer is either 500 or 2,000 and can't decide, typing 500 earns an average of about 550 points, typing 2,000 about 500, and typing the “safe” middle value of 1,000 only about 140. Splitting the difference is the worst choice here.
The real value of each round is the answer itself. Every time you see the true number, you gain a new anchor: now you know what a beaver weighs, or how long Lima to New York really is. Pay attention to the direction of your misses too. A common pattern is to pull every guess toward the middle, underestimating huge quantities and overestimating tiny ones. Once you notice your own bias, correcting for it is one of the fastest ways to improve.
Ready to try it? Pick a game from the main menu, or read how much everyday things really weigh to stock up on anchors first.