12 estimation jar ideas, with a plan for fair guesses
12 jar ideas · 15–20 minutes · counted objects and benchmark reasoning
By the Right Price team
An estimation jar works best when players can explain why they changed a guess. Use the material ideas below to prepare a jar, collect an initial estimate, reveal one useful benchmark and collect a revised estimate. The host counts the actual contents before play; there is no reliable universal answer for a particular jar size.
Choose a material and a useful benchmark
These are material choices, not twelve jars with predetermined answers. Pick two contrasting materials if time is short. Do not silently change the contents between guesses. A sample is only useful when its pieces resemble those in the jar.
| Material | What to notice | A benchmark to show |
|---|---|---|
| Large craft pom-poms | Gaps between soft shapes | Ten of the same pom-poms beside the jar |
| Wooden craft sticks | Orientation changes how tightly they pack | One bundle of ten matching sticks |
| Large building bricks | Different brick sizes change the estimate | Ten bricks of the same size as those inside |
| Paper clips | Overlapping shapes can hide pieces | A row of ten matching clips |
| Large buttons | Mixed sizes make equal layers unreliable | A small counted sample with the same size mix |
| Bottle caps | Nested caps can be hard to see | Five caps shown separately |
| Pencil-top erasers | Objects may sit behind the visible edge | Ten matching erasers in a clear cup |
| Corks | Horizontal and vertical corks pack differently | Five corks placed alongside the jar |
| Folded paper stars | Folds leave different amounts of empty space | A counted sample from the same batch |
| Large dried leaves | Size and curling affect the space used | Five leaves with a similar size range |
| Small pine cones | Shape leaves uneven gaps | Three comparable cones |
| Sealed counters | Identical flat pieces can form visible stacks | A stack of ten of the same counters |
Make a host count log before the game
- Give each jar an ID and record the material, size mix and container description. Count objects into groups of ten plus a remainder; a second adult or a second pass checks the total.
- Write first count, second count and agreed answer. If they disagree, recount before play. Seal the jar or keep it in sight so the answer stays fixed.
- Choose exactly one clue to reveal after first guesses: a counted comparison sample, a visible layer count, or an honest lower/upper bound you have checked.
- Prepare a player log with jar ID, first guess, revised guess and a sentence explaining the change. Keep the true-count column on the host sheet only.
Run a two-guess reveal
Give everyone the same viewing position and time; rotate the jar or show the same front and side photos to all teams. Collect one first estimate, reveal the benchmark and allow a revised estimate. Finally reveal and verify the actual count by grouped counting.
Use closest overall for the paper game: the smallest absolute difference from the true count wins, whether above or below. If two teams are equally close, share the win. Keep this score outside the app because native no-overbid judging uses a different rule. Discuss reasoning even when a revised guess is farther away; revision is not guaranteed to improve accuracy.
For a 15-minute session, spend three minutes explaining, eight minutes on two jars and four minutes comparing strategies. Participants may work in pairs or pass on a round. Do not use a hidden answer on this public page for a jar you have not counted.
Six benchmark practice cards with known facts
These short numeric cards teach the method before a real jar is shown. They provide all needed facts; their answers are calculations or scores, not claims about how many objects fit in an unseen container.
1. Groups of ten
The host counts seven full groups of ten counters and four loose counters. What is the verified total?
Host answer and explanation
74
7 × 10 + 4 = 74. Grouping makes a second count easier to audit.
2. An equal-layer model
In a simplified drawing, a jar has five layers with exactly eight blocks in every layer. How many blocks are shown?
Host answer and explanation
40
5 × 8 = 40. A real irregular jar may not have equal layers, so the stated model matters.
3. Distance from the answer
Your revised guess is 68 and the verified jar count is 74. How far away is the guess?
Host answer and explanation
6
The absolute difference is 74 − 68 = 6 objects.
4. An overestimate can be equally close
The true count is 74. One team guesses 68 and another guesses 80. How far is either guess from the answer?
Host answer and explanation
6
Both are six away. Under this closest-overall paper rule, the teams tie.
5. Did the revision help?
Your first guess is 60 and your revised guess is 70. The true count is 74. By how many objects did your error decrease?
Host answer and explanation
10
First error: 14. Revised error: 4. The decrease is 14 − 4 = 10.
6. Check a count log
A host records four groups of ten, three groups of five and two loose objects. What total should the log show?
Host answer and explanation
57
40 + 15 + 2 = 57. The log records a counted total; it is stronger evidence than judging jar volume.
Make the next round more informative
Keep the jar and contents fixed but offer a better comparison sample. Ask whether players used visible objects, layers, or a known group size. If a group assumes every layer is identical, turn the jar and show why uneven packing can make that model approximate.
For a number-focused classroom follow-up, compare the first and revised errors without naming a best or worst child. For a party, keep the explanation optional and share tied wins. You can reuse the containers with a newly counted fill on another day, but update the private key and any game answers first.
Use Guess a Number for your counted jars
- Count and check your own jars first. Open the game builder, sign in, enter a Game Title and choose Create from scratch.
- In Add Questions, write the jar ID and object description in Your Question, select Guess a Number, and enter your verified count in Correct Answer. Use a count, not a dollar price. Enter teaching notes manually; this article does not load a jar or its answer into the game.
- Show the physical jar or your own clear photographs and preview each reveal privately. Keep the first/revised-guess log and closest-overall scores on paper; the app does not automatically run this two-guess lesson or calculate improvement scores.
The public guide can be read without an account. Creating or hosting a game uses the product's sign-in and access flow; check current plans before your event. Answer disclosures on this page are convenient to hide, not private or access-controlled.