Recently, during this ARB trading competition, I reviewed my operations and want to share my experiences and lessons learned.

**Things Done Right**

Initially, I focused on $NIGHT trading, but I found the hype was too high, and the token's total supply was not stable enough, causing the wear costs to rise sharply. I responded quickly and immediately shifted to the ARB trading competition, which had similar incentives but much lower risk.

For $VSN, I carefully calculated the costs—most ordinary users face wear costs between 1/1000 and 2/1000. Based on this ratio, trading 10,000 dollars would burn about 10 U. I didn’t follow the usual approach; instead, I operated directly on-chain through my wallet. In the end, I reached a trading volume of 60,000 dollars, with wear controlled around 20 U, bringing the cost down to about 0.3%, which is a relatively ideal level.

Another key point was learning to use tools like the done data dashboard to predict cutoff scores, avoiding blind guesses.

**Things Done Wrong**

The biggest regret is missing out on the gains from $LAVA and $NIGHT. The logic is quite clear—these trading competitions push popularity to the peak, attracting a large influx of users who usually don’t trade, causing trading volume to surge and token prices to fluctuate accordingly. But I fell into a common trap: following the trend to mine profits, ignoring simpler ways to earn. Essentially, we should focus on earning from the miners, similar to the logic of selling shovels. In plain terms, holding some tokens at the right time and selling at the peak of hype is a more stable strategy.

The situation with NIGHT was even more outrageous—before confirming the profits, I rushed to hedge, which ended up causing losses. This is a typical rookie mistake of not pursuing certain gains.

Additionally, seeing too many people sharing ARB competition opportunities made me nervous about my trading volume, leading me to trade excessively. Only afterward did I realize that the volume I added and the actual effective trading volume had discrepancies, and some of my efforts didn’t pay off.