Crypto Price Prediction Methodology

Crypto markets move quickly and can be difficult to interpret. To help users better understand how prices may develop, Finst provides monthly crypto price predictions based on a transparent and consistent methodology.

These predictions are not financial advice, but they offer useful context by showing a prediction based on how a cryptocurrency has historically behaved, how market cycles influence performance, and how different scenarios might impact future prices.

This methodology explains how these predictions are determined, so users can interpret them clearly and use them alongside their own insights and investment strategy.

Please note that the predictions are updated daily based on the methodology. The methodology itself was last updated in November 2025.

Disclaimer: Investing in crypto-assets involves risk of losses. Prediction data is based on historical data and is provided for informational and educational purposes only. Prediction data may not be complete or accurate, and do not constitute any representation, warranty or any financial, investment or other form of advice by Finst. Future prices may differ significantly from the presented prediction. Before trading any crypto-asset, you should do your own research and evaluate your risk appetite. Finst is not responsible for any losses which you may incur from trading crypto-assets.

How the Methodology Works

The model combines historical monthly averages, three scenarios (neutral, bullish, and bearish), and market cycles such as the Bitcoin halving. This results in a price prediction that takes into account both seasonal influences and broader market dynamics.

Based on the Current Price

The model uses the cryptocurrency's closing price (the last price of the previous day) as input for all calculations, which means the predictions are updated daily.

For example, if Bitcoin closed at €80,000 on November 16, 2025, this price will be used as the starting point for the calculations on November 17, 2025 (the following day).

Historical Data Analysis

The model calculates the average monthly return for each of the 12 months based on the cryptocurrency's average monthly closing prices over the past five years. If a cryptocurrency is relatively new and has less than one year of historical data available, Bitcoin's historical performance is used instead. When sufficient data is available, the model uses up to five years of the asset's own history.

To ensure that more recent market conditions have a stronger influence on the analysis, the model applies an Exponentially Weighted Moving Average (EWMA). This technique, originally introduced by J.P. Morgan and widely used in financial risk modelling, assigns exponentially decreasing weights to older observations. As a result, recent price movements contribute more to the final estimate, while older data still plays a role but with less influence.

The degree to which newer data is prioritized is determined by a parameter called lambda (λ). In general, a higher λ gives more weight to older data, resulting in a smoother trend, while a lower λ makes the model more responsive to recent market changes. Because the cryptocurrency market is known for its volatility and fast-shifting conditions, EWMA allows the model to react appropriately without relying solely on short-term fluctuations.

By combining long-term seasonal patterns with a method that gives more weight to recent data, the model captures both the historical behavior of each month and the more up-to-date dynamics of the market.

Scenarios

Before introducing the three scenarios, it is important to understand where the adjustment factors come from. Both the scenario factors in section 3 and the market cycle factors in section 4 are based on the same long-term analysis of the total cryptocurrency market capitalization from 2013 to 2024.

By studying how the overall crypto market behaved during neutral, bullish and bearish years, the model derives realistic adjustment factors that can be applied to monthly historical averages.

The model presents three scenarios, each applying an adjustment factor to the historical monthly averages:

1. Neutral

The neutral scenario simply follows the historical averages without any additional adjustments.

2. Bullish

Historical data shows that during bullish market phases, the crypto market can experience significantly higher growth than average. However, such extreme peaks are not suitable for use in a monthly prediction model, as they can lead to unrealistic or unstable price predictions.

For that reason, the bullish scenario applies a moderate upward adjustment to the historical monthly trend. This adjustment strengthens gains and softens losses, creating a more optimistic outlook while keeping the prediction grounded and usable. The bullish scenario therefore reflects conditions in which the market performs better than usual, but without assuming the extreme peaks seen in past bull cycles.

3. Bearish

Conversely, historical data also shows that during bearish market phases, cryptocurrencies can experience substantial declines. To reflect these conditions, the bearish scenario applies a downward adjustment to the historical monthly trend. This reduces gains and amplifies losses, capturing the type of performance typically seen during prolonged market downturns. The bearish scenario therefore represents a more pessimistic outlook in which the asset performs below its long-term average

Effect of Market Cycles

In addition to historical averages and scenarios, the model also considers broader market cycles. These cycles are derived from the long-term growth pattern of the total cryptocurrency market capitalization, using data from 2013 to 2024. By analyzing this full period and extracting the average recurring pattern, the model extends this cycle into future years.

The Bitcoin halving plays a central role in shaping these market cycles, as it has historically had a major influence on overall crypto performance. Therefore, the years surrounding each halving are classified as bullish, bearish, or neutral:

In a bullish year, gains are strengthened, and losses are softened by applying a positive adjustment factor, reflecting stronger-than-average market behavior.
In a bearish year, gains are weakened, and losses are amplified by applying a negative adjustment factor, reflecting how the market typically behaves during bearish periods.

Since the Bitcoin halving occurs roughly every four years, the model applies the following cycle structure:

Year	Cycle Type
Halving year	Neutral
1 year post-halving	Bullish
2 years post-halving	Bearish
3 years post-halving	Neutral
Next halving year	Neutral

Future Decay Factor

To keep long-term predictions realistic, the model applies a future decay factor that gradually reduces the influence of both the scenario factor, and the market cycle factor the further into the future the prediction extends. Short-term market conditions can strongly impact near-term predictions, but their reliability decreases over longer horizons. The decay ensures that these short-term effects do not dominate predictions far into the future.

Why Use This Approach?

We use this approach because historical data provides the foundation, while market cycles and scenarios add strength to the predictions.

Historically grounded: Each prediction is based on five years of monthly return data, processed through an EWMA to give more weight to recent market behavior.
Seasonal influences: The crypto market follows cycles with recurring patterns; some months historically perform better than others.
Multiple scenarios: Each price prediction shows neutral, bullish, and bearish outcomes, each with its own factor on monthly growth.
Market cycle adjustments: Years around the Bitcoin halving are assigned specific cycle factors (bullish, bearish, or neutral) based on how the total crypto market has behaved across multiple halving cycles from 2013 to 2024.
Daily updates: The model updates predictions every day using the latest prices, ensuring that the predictions continuously adapt to current market movements.

Limitations

There are some limitations in the model to be aware of:

Based on historical data: The model relies on past price data, but the past does not guarantee future outcomes.
Limited data: Newer cryptocurrencies may lack sufficient data, requiring reliance on Bitcoin's history instead of their own.
No external factors: The model reflects growth/loss based on averages but does not account for external influences such as regulatory changes, hacks or major announcements.
Simplistic scenarios: The model only considers three scenarios (neutral, bullish, bearish) and does not include sideways trends or sudden recoveries after crashes.
Bitcoin-centric: The classification of neutral, bullish, and bearish years is based on Bitcoin halving, and does not account for possible altcoin seasons, where altcoins may outperform Bitcoin.
Bullish scenarios do not guarantee upward outcomes: Even in the bullish scenario, some cryptocurrencies may still show declining prices in the long-term forecast. This happens when the asset's historical performance over the past five years is predominantly negative. The bullish factor strengthens gains and softens losses, but it cannot reverse a consistently negative underlying trend.

Risks

Not investment advice: Finst's price predictions are only projections and should never be considered financial advice to buy or sell crypto.
Volatility: Since the crypto market can be extremely volatile, actual prices may diverge significantly from forecasts.
Personal responsibility: Each user is fully responsible for their own trading decisions. Finst's crypto price predictions are intended solely for illustration and educational purposes.
Model uncertainty: All prediction models rely on assumptions and historical patterns, which may not hold in the future. Unexpected market behavior can lead to outcomes far different from the projected values.
Long-term uncertainty: The further into the future a prediction extends, the greater the uncertainty. Long-term projections should be interpreted with extra caution.
External events: Sudden regulatory changes, exchange issues, hacks, macroeconomic events, or unforeseen market shocks are not incorporated into the model and can cause significant price movements.