From e25942b2c30eb1b4dee31049a7f9c42dc09b074a Mon Sep 17 00:00:00 2001
From: NunoSempere <nuno.semperelh@protonmail.com>
Date: Sat, 13 Apr 2024 22:14:59 -0400
Subject: [PATCH] tweak polls in main and restructure README

---
 README.md            | 55 ++++++++++++++++++++---------
 data/polls/fields.sh |  2 ++
 main.go              | 84 +++++++++++++++++++++++++++++++++++++++++++-
 3 files changed, 124 insertions(+), 17 deletions(-)
 create mode 100644 data/polls/fields.sh

diff --git a/README.md b/README.md
index b2d23c0..81e7ec3 100644
--- a/README.md
+++ b/README.md
@@ -1,19 +1,42 @@
-Electoral votes/election
+# Nuño's deads imple election simulator
 
-Year | Republican electoral college votes
-2000 | 271
-2004 | 286
-2008 | 173
-2012 | 206
-2016 | 304
-2020 | 232
+## First round: just consider the base rates.
 
-Year | Democrat electoral college votes
-2000 | 266
-2004 | 251
-2008 | 365
-2012 | 332
-2016 | 227
-2020 | 232
+- [x] Get past electoral college results since 2000
+- [x] Get number of electors for each state with the new census
+- [x] Combine the two to get an initial base rates analysis
 
-Ok, so because Obama really trounced the opposition, it won by more than the Republicans won, which means that the baserates approach didn't really work.
+This initial approach gives a 25% to republicans winning in the 2024 election. Why is this? Well, consider the number of electoral college votes: 
+
+| Year | Republican electoral college votes | 
+| ---- | --- | 
+| 2000 | 271 | 
+| 2004 | 286 |
+| 2008 | 173 |
+| 2012 | 206 |
+| 2016 | 304 |
+| 2020 | 232 |
+
+| Year | Democrat electoral college votes
+| ---- | --- | 
+| 2000 | 266 | 
+| 2004 | 251 | 
+| 2008 | 365 | 
+| 2012 | 332 | 
+| 2016 | 227 |
+| 2020 | 232 | 
+
+When Democrats won with Obama, they won by a lot, whereas when Republicans won with Bush and Trump, they won by a smaller amount. Or, in other words, this initial approach *doesn't take into account that states are correlated*. 
+
+Remedy: consider the conditional probabilities? But how? Or, relax assumptions using Laplace's law?
+
+## Second round: just consider polls
+
+- [ ] Download and format
+- [ ] Read
+- [ ] Consider what the standards error should be 
+- [ ] Aggregate polls?
+- [ ] Exclude polls older than one month?
+- [ ] ...
+
+## Third 
diff --git a/data/polls/fields.sh b/data/polls/fields.sh
new file mode 100644
index 0000000..99803b2
--- /dev/null
+++ b/data/polls/fields.sh
@@ -0,0 +1,2 @@
+
+cat president_polls_state.csv | head -n 2 | cut -d, -f13
diff --git a/main.go b/main.go
index 3d47891..2b9334e 100644
--- a/main.go
+++ b/main.go
@@ -14,6 +14,7 @@ type State struct {
 	Name                       string
 	Votes                      int
 	VictoriousPartyPerElection map[string]string
+	Polls                      []Poll
 }
 
 type VotesForEachParty struct {
@@ -21,6 +22,12 @@ type VotesForEachParty struct {
 	Republicans int
 }
 
+type Poll struct {
+	PollId      string
+	SampleSize  int
+	PollResults map[string]float64
+}
+
 // type src = *rand.Rand
 
 /* Globals */
@@ -87,6 +94,79 @@ func readStates() ([]State, error) {
 		results_file.Close()
 	}
 
+	/* Read polls */
+	polls_file, err := os.Open("data/polls/president_polls_state.csv") // Make sure to update this path
+	if err != nil {
+		return nil, fmt.Errorf("error opening the polls file: %v", err)
+	}
+	defer polls_file.Close()
+
+	// Using a temporary map to group poll results by state and poll ID
+	tmp_polls := make(map[string]map[string]Poll)
+
+	polls_reader := csv.NewReader(polls_file)
+	_, err = polls_reader.Read() // Skip the header
+	if err != nil {
+		return nil, fmt.Errorf("error reading polls header: %v", err)
+	}
+
+	for {
+		record, err := polls_reader.Read()
+		if err != nil {
+			break // EOF or an error
+		}
+		state_name := record[12]
+		// fmt.Printf("State: %s", state_name)
+		poll_id := record[0]
+
+		sampleSize, err := strconv.Atoi(record[22])
+		if err != nil {
+			continue // If error, skip this record
+		}
+
+		candidateName := record[44]
+		pct, err := strconv.ParseFloat(record[47], 64) // pct is in the 42nd column
+		if err != nil {
+			fmt.Printf("Error parsing percentage")
+			continue // If error, skip this record
+		}
+
+		if _, exists := tmp_polls[state_name]; !exists {
+			tmp_polls[state_name] = make(map[string]Poll)
+		}
+
+		poll, exists := tmp_polls[state_name][poll_id]
+		if !exists {
+			poll = Poll{
+				PollId:      poll_id,
+				SampleSize:  sampleSize,
+				PollResults: make(map[string]float64),
+			}
+		}
+		poll.PollResults[candidateName] = pct
+		tmp_polls[state_name][poll_id] = poll
+	}
+
+	// Add the aggregated poll data to the respective states
+	for state_name, polls := range tmp_polls {
+		var pollsSlice []Poll
+		for _, poll := range polls {
+			pollsSlice = append(pollsSlice, poll)
+		}
+		if state, exists := states[state_name]; exists {
+			state.Polls = pollsSlice
+			states[state_name] = state // Not redundant
+		} else {
+			// fmt.Printf("Encountered new state: %s\n", state_name)
+			/*
+				states[state_name] = State{
+					Name:  state_name,
+					Polls: pollsSlice,
+				}
+			*/
+		}
+	}
+
 	// Convert statesData map to a slice for returning
 	var states_slice []State
 	for _, state := range states {
@@ -97,6 +177,7 @@ func readStates() ([]State, error) {
 }
 
 func sampleFromState(state State) VotesForEachParty {
+	fmt.Printf("%s\n\n", state)
 	switch state.Name {
 	case "Nebraska":
 		// 2000: R
@@ -146,6 +227,7 @@ func simulateElection(states []State) int {
 
 	republican_seats := 0
 	for _, state := range states {
+		// fmt.Printf("%s\n", state)
 		election_sample := sampleFromState(state)
 		republican_seats += election_sample.Republicans
 		/*
@@ -176,7 +258,7 @@ func main() {
 		return
 	}
 
-	n_sims := 100_000
+	n_sims := 1
 
 	p_republicans := 0.0
 	for i := 0; i < n_sims; i++ {