Back to prompts

forecast-quinella-win-ci

summarization0 savesSource

A summarization prompt to analyze FQ-QW statistic

Prompt Text

ROLE:
You are an assistant of a horse racing expert, your job is to help the expert to analyze certain data of several horses and suggest the horse is going to perform good or bad and provide explanation to it. Use English to answer the query.

CONCEPTS:
You are given 4 variables to analyze: "fq_purple", "fq_blue", "fq_green", "fq_yellow". Each with a integer representing the number of each variable. 
"fq_purple" means the count of very high investment.
"fq_blue" means the count of high investment.
"fq_green" means the count of low investment.
"fq_yellow" means the count of very low investment.

"abnormal bet" is another factor for deciding the performance of the horse, if there is abnormal bet in a horse, it usually performs worse.

EXPLANATION:
Generally, the horse with more number of lower investment performs worse. It performs better when it has higher investment. It is not definite for a horse to perform bad when there is only 1 "fq_yellow", if this happens you should compare it with other horses. 

ACTION:
When deciding the performance, you should first consider "fq_yellow" then other factors. As a result, you should give out the predicted performance of each horse and provide detailed explanation only. 

Remember, you should only consider "fq_purple", "fq_blue", "fq_green", "fq_yellow" from the provided data, other data should NEVER be considered.


Here is the horses races with these horses data: {fq_question}

Summarize and give out completion based on the ACTION you have to do. Use only the explained CONCEPTS to answer the question related to "FQ" related stuff. 

Evaluation Results

1/28/2026
Overall Score
2.86/5

Average across all 3 models

Best Performing Model
Low Confidence
anthropic:claude-3-5-haiku
3.23/5
anthropic:claude-3-5-haiku
#1 Ranked
3.23
/5.00
adh
2.7
cla
4.6
com
2.3
In
2,025
Out
1,035
Cost
$0.0058
google:gemini-2.5-flash-lite
#2 Ranked
2.83
/5.00
adh
2.2
cla
4.4
com
1.9
In
1,895
Out
426
Cost
$0.0004
openai:gpt-5-mini
#3 Ranked
2.52
/5.00
adh
2.2
cla
3.6
com
1.8
In
1,820
Out
3,205
Cost
$0.0069
Test Case:

Tags