Crypto Price Prediction Tracker: Forecasts vs Actual Prices

Last Updated: October 30, 2026

Quick Answer: This tracker covers 48 of our coin pages, refreshed from October 24 to 28, 2026: it records the CoinCodex and CoinLore end-of-year forecasts on those pages and scores them against CoinMarketCap prices, misses included. A progress check of seven forecasts for the end of 2026 that stood on six of our pages in May and June found the October 30 price inside the published range for only one of them, Aave’s, less than 1% below the low of one of two Litecoin forecasts, and below the central figure of all seven, by 15% to 95%. Final scores for those forecasts, for the 88 model estimates now on record for 47 coins and for the end-of-2026 forecasts of our own that 11 coin pages state, retired models included, will be published in January 2027 and then updated every quarter.

Most of our price prediction pages do not set price targets of their own. They report forecasts from third-party models, and this page checks how those forecasts turn out, along with the few forecasts of our own that some pages state, using fixed rules that are set out below and do not change after the results are in.

Key Takeaways

• The tracker scores the forecasts our crypto price predictions publish, mainly from the CoinCodex and CoinLore models, together with the forecasts of our own that some pages state; it adds no new forecasts
• Progress check: of seven spring 2026 forecasts for the end of 2026, only Aave’s had the October 30 price inside its range; the other six sat entirely above it, Litecoin’s DigitalCoinPrice range by less than 1%
• The central figure of every spring forecast is above today’s price, with a median gap of about 65% ; the largest gaps so far are SafeMoon (the price is 95% below the forecast average), Tamadoge (90%) and Worldcoin (87%)
• Some old forecasts were far from the market even while they were on the page: Worldcoin’s average was about 5.6 times its June 4 price and SafeMoon’s about 5.8 times its June 3 price
• Our own forecasts are scored too: 11 of our coin pages state end-of-2026 forecasts of our own, and all are on record for January, including the models retired from the Avalanche, Bitcoin Cash and Pudgy Penguins pages on October 24, whose 2026 averages were about five to ten times the October 24 prices; the TRON page’s 2024 model is close to the market
• On record for December 31, 2026: 88 central estimates from CoinCodex and CoinLore for 47 coins, recorded October 24 to 28; 69 imply a fall, 15 a rise and four are roughly flat
• CoinLore’s base case is below or level with the recorded price for all 41 coins it covers, while CoinCodex expects a rise for 15 of 47
• For 12 coins, the two models point in opposite directions, so at least one of each pair will miss its direction call

How the Tracker Works

Which pages. This first edition records the third-party forecasts on the 48 coin pages listed below, refreshed from October 24 to 28, 2026. Coin pages whose forecasts were updated earlier in October, such as the XRP and Shiba Inu pages, are not in that record, but the forecasts of our own that they state are recorded in a separate table, with those on five of the pages refreshed on October 24. Under the rules below, a page’s forecasts are added only when the page is updated, with the date of that update.

What we record. For each coin page, we record the end-of-year forecasts of the two models that appear on most refreshed pages, CoinCodex and CoinLore: each model’s central estimate, its published low and high, the price the page showed and the date. For CoinCodex, the central estimate is the average of its December forecast; where a page gives only CoinCodex’s end-of-year figure, as the Bitgert and VeThor pages do, we record that figure. For CoinLore, it is the base case for the year, with the bear and bull cases as the low and high. For forecasts of our own, we record the figures exactly as the page states them, with the price the page gives and its date.

What does not count. Forecasts from other providers that only some pages use, such as Changelly and BeInCrypto, which are left out so that like is compared with like; bull-case targets from other sources quoted in a page’s text; headline targets that sit at the top or bottom of a model’s range; and forecasts that a page mentions but does not use, for example because the provider’s page was stale or showed placeholder values. Forecasts of our own do count: they are scored in a separate table, under the rules set out with it.

Prices. Final scores use the CoinMarketCap daily close for December 31 (UTC). When CoinMarketCap no longer tracks a coin, as with Tamadoge, we use the most liquid market we can find and name it. If no market price can be found for December 31, the forecast is listed as not scorable, with the reason.

Scoring rules:

• Hit or miss: a forecast is a hit if the December 31 close falls inside the model’s published low and high. When a model gives a single figure, or when either end of its range is less than 10% from its central estimate, that end is moved out to 10%, so the band is never narrower than 10% either side. Both ends of every CoinCodex range on record are within 10% of its central estimate, and three CoinCodex entries are single figures, so every CoinCodex band is 10% either side of its central estimate. For CoinCodex, we also report whether the close fell inside its own published range, where it published one.
• Error: the percentage difference between the December 31 close and the central estimate. When a source gave no average, we use the midpoint of its low and high and say so.
• Direction: a central estimate above the price when recorded is a call for a rise, one below it a call for a fall. Estimates within 2% of that price count as flat and are not scored for direction. A call is right if the December 31 close moved the same way from the recorded price. For the spring 2026 forecasts, the recorded price is the CoinMarketCap close on the date of the saved version; Tamadoge’s spring forecast has no such price and is not scored for direction.
• Misses stay in: every recorded forecast appears in the results, including those that miss by wide margins.
• No edits after the fact: once recorded, a forecast is not changed. When a coin page updates its forecast, the new figures are added as a new entry with their own date.
• Schedule: final scores for the end-of-2026 forecasts in January 2027, then a quarterly update of the record.

A wide range is easy to hit. CoinLore’s bear cases on record sit a median of about 63% below the recorded price and its bull cases a median of about 3.3 times it, so for CoinLore the error from the central estimate and the direction call are the stricter tests.

What we could not recover. We had planned to score forecasts from 2025 as well, but archived copies of the older pages could not be retrieved, and our publishing system keeps only recent revisions. The earliest forecasts we can check are those that stood on the pages in spring 2026, before the October refresh. We recovered end-2026 forecasts from six saved versions; other saved versions we checked did not contain a 2026 forecast table we could match, and we did not reconstruct forecasts we could not read. We have not yet checked the saved versions of the pages refreshed on October 24, of the Bitgert and VeThor pages, or of the AIOZ Network, Dogelon Mars, Gala, IoTeX, Mog Coin, Quant, Spell Token, The Graph, Turbo and XYO Network pages.

Progress Check: Spring 2026 Forecasts vs Prices Today

Before updating each page in October, the editor saved its full text. Six of the saved versions we have checked, dated May 11 to June 4, 2026, contain a forecast table with a 2026 row. The figures below are copied exactly from those rows. With about two months left in the year, this is a progress check, not a final score.

Prices on October 30 are from CoinMarketCap at 09:57 UTC, except Tamadoge, whose latest available price is from October 28 (see the notes below). “Price then” is the CoinMarketCap daily close on the date of the saved version. Litecoin’s Changelly range gave no average, so the comparison uses its midpoint of $160.27.

Only one of the seven forecasts has the price inside its range so far: Aave’s, with the price about 31% above its low. Litecoin’s DigitalCoinPrice range still held the price on October 28, but by October 30 Litecoin was about 0.4% below its $55.59 low, so that result could change again before December 31. The central figures of all seven are above the October 30 price, with a median gap of about 65%. Every miss so far is a forecast that was too high, and the Worldcoin, Tamadoge and SafeMoon forecasts are off by 87% to 95%. Measured from the price on the date of each saved version, the direction call is right so far for three of the six forecasts that can be checked: all six called for gains, and Aave and Litecoin, with two forecasts, rose, while JasmyCoin, Worldcoin and SafeMoon fell. Some forecasts were far from the market even while they were on the page: Worldcoin’s average was about 5.6 times its June 4 price, SafeMoon’s about 5.8 times its June 3 price, and Changelly’s Litecoin low about three times the June 3 price.

Notes on this data:

• Sources: the old tables for Aave, Tamadoge and JasmyCoin named no source in the forecast row, and we could not confirm whether those figures came from a third-party model, so we show them as forecasts the pages published.
• Dates: the version date is the last change to each page before October, so the forecasts may be older. SafeMoon’s figures match the prices CoinMarketCap recorded for SFM in late 2025, when it closed inside that range on several days between October 29 and November 8, and are far above its June 2026 price, which suggests they were written well before the June version. CoinMarketCap tracks SFM as SafeMoon V2, now a Solana token, and the comparison uses that series.
• Tamadoge: CoinMarketCap lists TAMA as untracked and has recorded no price since December 24, 2025, so its price comes from the only active market we found, a Uniswap pool holding about $290, as read for our Tamadoge page on October 28 at about 10:05 UTC; later reads that afternoon and on October 30 at about 10:00 UTC returned no pool data, so the October 28 figure is kept. The result does not hinge on the exact figure: the forecast’s low was about five times that price and about 2.2 times CoinMarketCap’s last recorded price of $0.0000864, from December 24, 2025.
• Litecoin: the old page gave DigitalCoinPrice’s average only as an approximate figure, about $65, so that comparison is approximate too.

Forecasts on Record for End of 2026

The 48 coin pages in this record, refreshed from October 24 to 28, 2026, published these end-of-2026 central estimates. Each will be scored against the December 31 close under the rules above. Prices are as shown on each page, rounded as published, so an implied change can differ by about a point from the page’s own text.

SafeMoon is listed because its page publishes no forecast: the one current model we found for SFM, CoinDataFlow’s, is not used, because its forecasts are almost the same multiples of the price as for other coins. Tamadoge’s price is from the only active market we found, a Uniswap pool, as read at about 10:05 UTC. The Bitgert and VeThor pages give CoinCodex’s end-of-2026 figure rather than a December range, and the Hedera page gives its December average without a range, so those three entries are scored against a band of 10% either side. The TRON page gives CoinLore’s figures in its text, rounded to the cent.

Of the 88 central estimates on record for 47 coins, 69 imply a fall by the end of 2026, 15 a rise and four are within 2% of the recorded price: Beldex’s CoinLore base case and CoinCodex’s figures for AIOZ Network, Internet Computer and Tezos. CoinLore’s base case is below the recorded price for 40 of the 41 coins it covers and roughly level for Beldex, while CoinCodex expects a rise for 15 of 47 coins, led by Render (+182%), Hedera (+139%) and JasmyCoin (+103%). For 12 coins, Aave, Bitcoin Cash, Chainlink, Fetch.ai, JasmyCoin, Litecoin, Quant, Render, Solana, TRON, XDC Network and XYO Network, the two models point in opposite directions. The contrast with the spring forecasts is sharp: their central figures all sit above today’s prices, while most of the current estimates expect declines.

Our Own Models on Record for End of 2026

Some of our coin pages also state end-of-2026 forecasts of our own: older models that were retired on October 24, a 2024 model that the TRON page keeps for comparison, and the scenarios set out on the Arbitrum, Bitcoin, Cronos, Dogecoin, Ethereum, Shiba Inu and XRP pages. They are recorded below exactly as the pages state them and will be scored in January 2027 together with the third-party forecasts. They are kept in a separate table, so the counts above, 88 estimates for 47 coins, cover CoinCodex and CoinLore only.

Prices are as shown on each page, rounded as published, so an implied change can differ by a few points from the page’s own text. The Arbitrum and Shiba Inu pages give no central figure for their base case, so the implied change in brackets uses the midpoint of the range.

How these rows are scored. In January 2027, each row is scored against the CoinMarketCap close for December 31, 2026, in the same way as the third-party rows: a range counts as a hit if the close falls inside it, a single figure gets a band of 10% either side, and we report the error from the central figure and whether the direction call was right. Retired models are scored too, because they were published. Where a page sets out scenarios, the base case is the forecast we score, with the error measured from its stated center or, when none is stated, from the midpoint of its range, and we also report which scenario, if any, the close fell in. The Cronos page now states only the bear case of its April scenarios, so for Cronos we report only whether the close falls inside $0.040 to $0.070. The Avalanche, Bitcoin Cash, Pudgy Penguins and TRON models give a 2026 average, not a year-end figure, so we compare that average with the December 31 close as stated and say so in the results. Both ends of TRON’s 2026 range are within 10% of its average, so, as with CoinCodex, it is scored against a band of 10% either side, and we also report whether the close fell inside the range as published.

The retired Avalanche, Bitcoin Cash and Pudgy Penguins models put the 2026 average at roughly seven, five and ten times the October 24 prices of those coins, the day they were retired. The TRON model’s 2026 average of $0.328 is within 1% of the October 24 price of about $0.329. The Cronos page says that since April, CRO has mostly traded in the bear case of the scenarios it published then. The base cases of the six current scenario sets, for Arbitrum, Bitcoin, Dogecoin, Ethereum, Shiba Inu and XRP, all contain the price each page recorded; the Bitcoin and Ethereum ranges are centered about 1% to 2% above it, the Dogecoin, Shiba Inu and XRP ranges about 9% to 14% above it, and Arbitrum’s about 3% below.

What We Changed in October 2026

In October 2026, we rewrote the price prediction pages listed above. The 11 pages refreshed on October 24 use a shorter format: they compare named, dated forecasts year by year, and the Avalanche, Bitcoin Cash, Cronos, Pudgy Penguins and TRON pages explain how the older forecasts of our own that they carried compare with the market. The 37 pages refreshed from October 26 to 28 share a common format:

• Named models with dates: each forecast table names its source and each page dates its forecasts; many pages also give the date of each model’s page and check that the price it starts from is close to the market price
• Central estimates and ranges: for each year, the page shows each model’s low, average and high, or CoinLore’s bear, base and bull cases, the implied change from today’s price and, where a model’s headline target differs from its central estimate, whether that target is the top or the bottom of the model’s range
• Excluded providers, with reasons: pages leave out forecasts that start from a price far from the market, show placeholders such as a $0.00 price or unfilled template fields, or apply almost the same multiples of the current price to every coin, and they say which providers were left out and why
• No range without a usable model: when no current forecast passes these checks, the page says so and gives no range, as on the SafeMoon page; when only one current model passes them, as for Amp, Tamadoge and Turbo, the page uses only that one
• Price history and market data: each page adds a key data table and a price history table from CoinMarketCap, with year-end closes and the all-time high and low, and works out the market cap that a price target would imply
• Risks and sources: each page lists the main risks, names its data sources with a date and states that forecasts are not investment advice

The old tables were different: three of the six we could recover named no source for their 2026 forecast, and two carried averages more than five times the coin’s price on the date of the saved version.

How to Read Any Crypto Price Prediction

• Check the date and the starting price. A forecast built on an old price is stale, however precise it looks. Compare the price the forecast starts from with a live price on CoinMarketCap or CoinGecko.
• Find the central estimate. Headline targets are often the top or bottom of a range, not the middle. On our Aave page, for example, CoinCodex’s headline target of $184.08 is the bottom of its December range.
• Look at the width of the range. A range from about 60% below to three times today’s price covers most outcomes and says little; a narrow range is a stronger claim and easier to miss.
• Compare more than one model. When models disagree on direction, as they do for 12 of the coins above, no single forecast is a strong signal.
• Watch for templates. If a provider’s forecasts are the same multiples of the current price for every coin, they reflect nothing specific to that coin.
• Do the market cap math. Multiply a target price by the circulating supply and compare the result with today’s market cap and with larger coins.
• Remember what models miss. Most models extrapolate past prices; they do not anticipate delistings, token unlocks, exploits or court rulings. Check market-moving claims against primary sources; our guide on how to verify crypto news and rumors explains how.

Nothing on this page is investment advice. Cryptocurrency prices are volatile, and a model’s past accuracy does not guarantee its future accuracy.

Frequently Asked Questions

What is the crypto price prediction tracker?
It is a public record of the price forecasts published by 48 of our coin pages, scored against what prices do. For each coin, we log the central estimate and range of CoinCodex and CoinLore, with the price and the date, plus the forecasts of our own that 11 pages state. After the year ends, we compare them with the CoinMarketCap close and publish hits and misses alike.

How are the forecasts scored?
A forecast is a hit if the December 31 close falls inside the model’s published range, or within 10% of its central estimate when the range is narrower than that. We also report the percentage error from the central estimate and whether the forecast called the direction correctly. Misses stay in the record, and recorded forecasts are never edited after the fact.

How accurate were the spring 2026 forecasts so far?
Not very, with two months to go. Of seven end-of-2026 forecasts for six coins that stood on our pages in May and June, the October 30 price was inside the range for one, Aave’s, while one Litecoin forecast had just slipped below its low. All seven central figures exceeded the price, by a median of about 65%, and the SafeMoon, Tamadoge and Worldcoin forecasts were 87% to 95% too high.

Why do forecasts for the same coin differ so much?
The models use different methods and assumptions. CoinCodex’s December figures follow a narrow monthly path, while CoinLore publishes bear, base and bull cases that typically run from about 60% below the current price to more than three times it. For 12 coins on record, including Render and Solana, the two models expect opposite directions by the end of 2026, so at least one will miss.

Which forecast models does the tracker record?
For the end of 2026, it records CoinCodex, whose central estimate is its December 2026 average or, on two pages, its end-of-year figure, and CoinLore, whose central estimate is its base case. It also records the forecasts of our own that 11 coin pages state, retired models included. Other providers that only some pages use are not recorded, and neither are forecasts that a page reports but does not use.

When will the final scores be published?
Final scores for the end-of-2026 forecasts will be published in January 2027, based on CoinMarketCap’s daily close for December 31, 2026. After that, the tracker will be updated every quarter, adding new forecasts as the coin pages are refreshed and keeping every earlier entry, whether it was a hit or a miss, with its original date and figures.

Do your pages make their own price predictions?
Mostly not. Our coin pages mainly report forecasts from named third-party models and explain how they were built, when they were updated and why some providers are left out. A few also show scenarios or older models of our own, labelled as ours, as on the XRP and TRON pages. The tracker scores those too, alongside the third-party forecasts, in January 2027. Nothing here is investment advice.

Data sources: CoinMarketCap (prices on October 30, 2026, at 09:57 UTC and daily closes), DexScreener (Tamadoge’s Uniswap pool, as read on October 28, 2026), the forecast tables of our coin pages as published October 24 to 28, 2026, citing CoinCodex and CoinLore, the forecasts of our own stated on the Arbitrum, Avalanche, Bitcoin, Bitcoin Cash, Cronos, Dogecoin, Ethereum, Pudgy Penguins, Shiba Inu, TRON and XRP pages, as read on October 24 to 28, 2026, and the saved pre-update versions of six coin pages dated May 11 to June 4, 2026, citing DigitalCoinPrice, Changelly and Coinpedia where a source was named.