The ranges of estimates are important in terms of market reaction because when the actual data deviates from the expectations, it creates a surprise effect. Another important input in market's reaction is the distribution of forecasts.In fact, although we can have a range of estimates, most forecasts might be clustered on the upper bound of the range, so even if the data comes out inside the range of estimates but on the lower bound of the range, it can still create a surprise effect.Non-Farm Payrolls35K to 125K range of estimates50K-75K range most clustered65K consensusUnemployment Rate4.3% (8%)4.2% (74%) - consensus4.1% (18%)Average Hourly Earnings Y/Y3.0% (14%)2.9% (66%) - consensus2.8% (20%)Average Hourly Earnings M/M0.4% (7%)0.3% (72%) - consensus0.2% (21%)The reaction to data is going to be tricky because there could be something for everyone today. A bad report could strengthen growth concerns and extend the risk-off sentiment. At the same time, it could also lead to a dovish repricing in interest rate expectations and support risk assets on a less hawkish Fed. In the current context, I would say the former is more probable.The same goes for a good report. In this case, we might see a hawkish repricing that could weigh further on risk assets. But, good data could also trigger a relief rally on easing growth concerns. I guess the market would like to see good data, just not too hot. This article was written by Giuseppe Dellamotta at investinglive.com.