176th
Accuracy Rank

fsheng

About:
Show more
Forecasting Activity
Forecasting Calendar
 

Past Week Past Month Past Year This Season All Time
Forecasts 0 0 47 47 47
Comments 0 0 52 52 52
Questions Forecasted 0 0 22 22 22
Upvotes on Comments By This User 0 0 12 12 12
 Definitions
New Prediction
fsheng
made their 6th forecast (view all):
Probability
Answer
0% (0%)
Less than or equal to 1349
10% (-80%)
Between 1350 and 1499, inclusive
90% (+80%)
Between 1500 and 1649, inclusive
0% (0%)
Between 1650 and 1799, inclusive
0% (0%)
More than or equal to 1800
Why do you think you're right?

Revising my forecast based on today's CDC report. As of 9/17, a total of 1491 cases have been reported. This revision was based on a few things demonstrated by today's CDC data: 

  1. There was a significant lag in case reporting because the data was not real-time.  Data already reported should not be treated as static, rather "living," and delayed reported cases will be accumulated over time.
  2. My last week's forecast was based on reporting last week, which showed significantly fewer incidents per week, which led to an underestimation of incidents/week (I used a 10-15 range per week last week, but apparently my upper bound was lower than what today's data shows for early Sept).
  3. With a nearly 2-week window,  an additional 9 should be very likely.
  4. Sometimes, there are benefits in sitting on the data for a while, instead of reacting immediately. Especially in cases like this. Incomplete data and delayed reporting could be misleading and misinforming.
Files
Why might you be wrong?

1. Ongoing data tracking/reporting could be disturbed/disrupted by turmoil related to mass personnel exodus and turnover, which compromised the data updates.

2. Extreme weather and sudden temporary drop might happen (the chance is not zero), which will affect the virus survival and spread, leading to significantly fewer new cases.

Files
New Prediction
fsheng
made their 3rd forecast (view all):
Probability
Answer
95% (+15%)
Less than or equal to 1
5% (-15%)
Between 2 and 3, inclusive
0% (0%)
Between 4 and 5, inclusive
0% (0%)
More than or equal to 6
Why do you think you're right?

With three weeks left, I think the chance of having >1  major hurricane becomes very small. Therefore, I'm going to increase my confidence in previous forecasts, which seems to be corroborated by crowd wisdom.

Files
Why might you be wrong?

Leaving wiggle room for uncertainty seems wise and scientifically sound.

Files
New Prediction
fsheng
made their 2nd forecast (view all):
Probability
Answer
80% (0%)
Less than or equal to 1
20% (0%)
Between 2 and 3, inclusive
0% (0%)
Between 4 and 5, inclusive
0% (0%)
More than or equal to 6
Confirmed previous forecast
Files
New Prediction
Confirmed previous forecast
Files
New Prediction
Confirmed previous forecast
Files
New Prediction
New Badge
fsheng
earned a new badge:

Active Forecaster

New Prediction
New Badge
fsheng
earned a new badge:

Star Commenter - Aug 2025

Earned for making 5+ comments in a month (rationales not included).
New Prediction
fsheng
made their 3rd forecast (view all):
Probability
Answer
0% (0%)
Less than or equal to 500
85% (+75%)
Between 501 and 750, inclusive
15% (-75%)
Between 751 and 1000, inclusive
0% (0%)
Between 1001 and 1250, inclusive
0% (0%)
More than or equal to 1251
Why do you think you're right?
Based on the data from 7/24-8/22, the daily average is 24.37.   Based on the data from 8/15-822, the daily average is 22.67. Therefore, total monthly events are estimated to range between 703-755. The monthly data do not reveal clear upward/downward trend. With 3 days left for this month, the likelihood of totaling below 750 is slightly higher,
Files
Why might you be wrong?

Depends on what the data looks like for the last week of Aug, the total number may be above 750, especially if there is more than average political violence happening in the last week.

Files
Files
Tip: Mention someone by typing @username