PumpView Wash Score Guide: Spot Fake Volume on Solana DEXes
Wash trading is one of the biggest problems in Solana memecoins and low‑cap tokens. Bots loop trades between their own wallets to fake volume, push tokens into “trending” lists, and lure in real buyers.
PumpView’s Wash Score (0–100%) is built specifically to help you see through that noise.
This guide explains:
- What wash trading looks like on Solana DEXes
- How PumpView’s wash score is actually computed
- How to read the score in Hot Tokens and Early Scanner
- Practical trading rules of thumb using wash score
- Limitations and false positives you should understand
Everything here is grounded in real on‑chain research on DEX wash trading and PumpView’s own public documentation.
What Is Wash Trading on Solana DEXes?
Wash trading is when the same entity trades with itself (or a tight cluster of its own wallets) to create fake volume and misleading price action. On Solana, this is usually done on AMM DEXes like Raydium and launchpads like Pump.fun / PumpSwap by:
- Rapid back‑and‑forth swaps between the same wallets
- Coordinated loops across a few wallets controlled by one actor
- Bursts of tiny trades to inflate “trades count” and volume
On centralized exchanges, wash trading is often hidden in opaque order books. On DEXes, it’s all on‑chain, but you need the right heuristics to detect it.
Recent research confirms that wash trading exists on DEXes, even if it’s a small share of total volume at the ecosystem level. Chainalysis, for example, estimates that suspected wash trading accounted for a fraction of a percent of total DEX volume in late 2024, but still hundreds of millions of dollars in absolute terms. (chainalysis.com)
On Solana specifically, multiple analyses and news reports have highlighted that parts of the DEX volume—especially in low‑liquidity memecoins—are likely driven by bots and wash trading to manipulate metrics and attract traders. (coinglass.com)
For you as a trader, the key takeaway is simple:
High volume and fast candles do not always mean real demand.
You need a way to separate organic trading from manufactured churn. That’s what PumpView’s wash score is for.
How PumpView’s Wash Score Works (High‑Level)
PumpView’s public feature description explains that every token in Hot Tokens gets a Wash Score (0–100%) based on four independent signals computed from the last 60 seconds of trades. (pumpview.fun)
The goal is to estimate what share of recent activity looks like wash trading, not to prove fraud in a legal sense. It’s a real‑time heuristic, not a courtroom standard.
60‑Second Rolling Window
PumpView ingests live swaps from major Solana DEX venues:
- Pump.fun / PumpSwap
- Raydium (CPMM & CLMM)
- Meteora (including DYN pools)
- Other supported Solana DEXes and launchpads
Each token’s recent trades (roughly the last 60 seconds) are analyzed to compute the wash score. This short window makes the score responsive to current behavior instead of being anchored to old data. (pumpview.fun)
Four‑Signal Fake Volume Analysis
PumpView’s docs describe a “four‑signal fake volume analysis”. One of the signals is explicitly mentioned:
- Same‑wallet round trips (40% weight)
- Detects wallets that both buy and sell the same token within the 60‑second window.
- Measures their share of total volume in that window.
- If a few wallets are responsible for a large fraction of both buys and sells, that’s a strong wash‑trading pattern. (pumpview.fun)
The other three signals are not fully detailed in the public copy, but they follow the same principle as academic and industry DEX wash‑trading heuristics:
- Look for self‑trading and repeated trading loops between the same addresses
- Flag abnormally dense trade patterns that don’t match normal user behavior
- Use wallet‑level and token‑level context to distinguish organic flow from manipulation (coincub.com)
PumpView combines these four signals into a single Wash Score from 0% to 100%:
- 0% ≈ no detectable wash‑like patterns in the last minute
- 100% ≈ almost all recent activity matches wash‑like patterns
The exact formula and weights beyond the 40% same‑wallet component are proprietary, but the behavior is observable in the UI: tokens with clear bot churn and self‑loops tend to show high scores; tokens with many unique traders and organic flow tend to show low scores.
Where You See Wash Score in PumpView
1. Hot Tokens Table
In Hot Tokens, each row (token) includes:
- Buy Score (0–12): PumpView’s multi‑factor momentum score
- Wash Score (0–100%): estimated share of fake / wash‑like activity
- Other metrics like 1m/5m candles, DEX venue, and more (pumpview.fun)
Here, wash score helps you filter which “hot” tokens are actually tradeable:
- A token can have a strong Buy Score but a dangerously high Wash Score – meaning the momentum might be fake.
- Another token might have a moderate Buy Score but very low Wash Score, suggesting cleaner, organic flow.
2. Early Scanner Bubbles
In Early Scanner, new tokens appear as bubbles sized and colored by activity. Wash score is available in the token details panel when you click a bubble.
This is especially important for brand‑new Pump.fun / PumpSwap launches, where:
- Devs and bot operators often spam trades to hit trending pages on tools like DexScreener or Birdeye. (reddit.com)
- You have almost no historical data, so real‑time wash detection is one of the few objective signals you have.
How to Interpret Different Wash Score Ranges
PumpView doesn’t impose hard rules on what you should or shouldn’t trade, but you can build your own framework around these ranges.
0–15%: Clean Flow Zone
- What it usually means
- Most recent volume comes from distinct wallets trading in a natural pattern.
- Little evidence of the same wallet (or tight cluster) sitting on both sides of the market.
- How to use it
- Good baseline for “organic” tokens.
- If Buy Score is strong and wash score is this low, the move is more likely driven by real demand.
15–35%: Mildly Noisy but Often Tradeable
- What it usually means
- Some bot or arbitrage activity is present.
- You might see a few wallets doing both buys and sells, but not dominating the book.
- How to use it
- Treat as yellow zone: do extra checks on Birdeye / DexScreener (liquidity, unique traders, spread).
- Fine for many short‑term trades, but don’t assume all volume is clean.
35–60%: Heavy Bot / Wash Risk
- What it usually means
- A significant chunk of recent volume looks like round trips or tight wallet clusters.
- High chance that someone is manufacturing volume to hit trending lists or bait entries.
- How to use it
- Only touch if you understand the risk and have a clear plan (tight stops, small size).
- Cross‑check:
- Liquidity vs volume (e.g., 5 SOL liquidity but 200 SOL 5‑min volume is a red flag). (webopedia.com)
- Number of unique traders on Birdeye / DexScreener.
60–100%: Mostly Manufactured Volume
- What it usually means
- Almost all recent activity matches wash‑like patterns: same wallets buying and selling, loops, or dense bot churn.
- Classic “volume bot” behavior seen in many Solana memecoins that spike on trackers then rug or fade. (reddit.com)
- How to use it
- For most traders, this is a do‑not‑touch zone.
- If you trade it at all, treat it as pure gambling and size accordingly.
Remember: these ranges are heuristics, not guarantees. A token can have a temporarily high wash score during a bot‑driven arbitrage burst and then normalize later.
Practical Trading Playbook Using PumpView Wash Score
Here’s a concrete way to integrate wash score into your Solana DEX trading.
Step 1: Start in Hot Tokens, Sort by Buy Score
- Open Hot Tokens.
- Sort by Buy Score (descending).
- For the top candidates, immediately check Wash Score.
Rule of thumb:
- Skip tokens with Wash Score > 60%, unless you have a very specific reason.
- Prioritize tokens with Buy Score high + Wash Score < 25%.
Step 2: Cross‑Check with External Tools
For any token you’re about to trade, combine PumpView with:
- Birdeye – check:
- Liquidity vs 5m/1h volume
- Unique traders and holders
- Price impact for your trade size
- DexScreener – check:
- Candles across multiple DEX pools (Raydium, Meteora, etc.)
- Spread and depth
- Solscan / SolanaFM – check:
- Dev wallet behavior (huge supply still in one wallet, recent suspicious transfers)
If PumpView shows low wash score and these tools show healthy liquidity + many unique traders, odds are higher that the move is real.
Step 3: Use Wash Score in Early Scanner for New Launches
When hunting new Pump.fun / PumpSwap launches in Early Scanner:
- Watch for new bubbles with rising Buy Score.
- Before aping, open the token details and check Wash Score.
You can build simple rules like:
- Avoid first entries if wash score is > 50% in the first minutes.
- Prefer new tokens where:
- Wash score stays < 25% even as volume spikes.
- Multiple DEX venues (Raydium + Meteora, etc.) start to show organic flow.
This helps you avoid becoming exit liquidity for volume bots that never intended to build a real market.
Step 4: Monitor Wash Score While You’re In a Trade
Wash score is dynamic. Even if it was low when you entered, it can spike later.
Practical uses:
- If you’re in profit and see wash score suddenly jump from, say, 10% → 55% while price stalls:
- Consider taking partial profits or tightening your stop.
- This can signal that organic buyers are fading and bots are taking over.
- If you’re down and wash score climbs high:
- That’s a sign the move may have been manufactured from the start.
- Don’t rely on “volume will come back” – it might never have been real.
You can also combine this with PumpView’s custom signal strategies and alerts, e.g., alert when a token you watch crosses a certain wash score threshold while volume or Buy Score changes.
Limitations and False Positives You Should Know
No wash‑trading detector is perfect. Academic work on DEX wash trading and industry tools like Bitquery’s Solana wash‑trade detector all highlight the same challenges: complex MEV, arbitrage, and routing can look similar to wash trading at first glance. (coincub.com)
For PumpView’s wash score, keep these caveats in mind:
- Short‑Window Focus
- It looks at roughly the last 60 seconds of trades.
-
A brief arbitrage burst can temporarily spike the score even if the token is healthy overall.
-
Arbitrage & MEV Bots
- Legitimate arbitrage between Raydium, Meteora, and other DEXes can create dense, repetitive patterns.
-
Some of these may be flagged as wash‑like, even though the intent is price alignment, not fake volume.
-
Clustered Retail Behavior
- In very early memecoins with tiny communities, a few real wallets might dominate volume simply because there are few participants.
-
This can look similar to a wash cluster, especially when liquidity is low.
-
Not a Rug‑Pull Detector
- A token can have low wash score and still rug via liquidity removal or mint authority abuse.
- Use separate tools and checks (e.g., RugCheck, Solscan, contract audits) for rug‑pull risk.
The right mindset is: wash score is a strong filter, not a guarantee. It helps you avoid the most obviously manufactured volume, but it doesn’t replace broader due diligence.
Putting It All Together
PumpView’s wash score gives Solana traders a real‑time, token‑level view of how much recent activity looks like wash trading.
- It’s computed from four independent signals over the last 60 seconds of trades, with same‑wallet round trips as a key component. (pumpview.fun)
- You see it directly in Hot Tokens and Early Scanner, right next to Buy Score and other metrics.
- Used correctly, it helps you:
- Filter out fake “hot” tokens driven by volume bots
- Prioritize tokens with organic flow and real demand
- React faster when a token you’re in starts to show manufactured churn
To get the most out of it:
- Build personal thresholds (e.g., rarely trade tokens with wash score > 60%).
- Always cross‑check with Birdeye, DexScreener, and Solscan.
- Treat wash score as a risk lens, not a single yes/no switch.
In a Solana environment where bots can spin up millions of fake trades cheaply, having a transparent, on‑chain‑driven wash score is a real edge. Use it as a core part of your process whenever you’re scanning Hot Tokens or sniping new launches on Solana DEXes.